1 Department of Anaesthesia and 2 Söder Hospital, Karolinska Institute, S-118 83 Stockholm, Sweden
*Corresponding author. E-mail robert.hahn{at}SOS.SLL.SE
Accepted for publication: November 4, 2003
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Twelve patients aged between 27 and 51 (mean 40) underwent laparoscopic cholecystectomy. An i.v. infusion of 1.4 litres of glucose 2.5% over 60 min was started when surgery began. A volume kinetic model was fitted to measurements of the plasma glucose concentration and the degree of haemodilution. Nomograms were constructed based on the kinetic results.
Results. The volume of distribution for the glucose and infused fluid and the plasma insulin levels were similar to the ones recorded in previous volunteer studies, but 5070% lower values were obtained for the clearance of glucose (mean 0.21 litres min1), endogenous glucose production (1.1 mmol min1) and the elimination rate constant for the infused fluid (median 37 ml min1). Urinary excretion was markedly depressed and amounted to 9% of the infused fluid volume 4 h after starting surgery. To prevent hyperglycaemia, nomograms suggested that the infusion should be directed towards a target glucose concentration and then slowed down in a controlled way. At steady state, the infused fluid maintains a 3.5% plasma dilution for each mmol that plasma glucose remains above baseline.
Conclusion. Metabolic changes warrant careful balancing of infusion rates of glucose 2.5% during laparoscopic cholecystectomy, which is facilitated by a nomogram. Volume expansion from the infused fluid volume should be recognized.
Br J Anaesth 2004; 92: 48592
Keywords: blood, haemodilution; fluids, i.v.; metabolism, glucose; pharmacokinetics
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The administration of glucose during surgery has been considered questionable as trauma-induced release of catecholamines, cortisol, and growth hormone impedes the glucose-lowering effect of insulin, which leads to hyperglycaemia.4 The metabolic response is, however, reduced using laparoscopic techniques as compared with open surgery.5 6 Many clinicians use glucose 2.5% with electrolytes during surgery, which is essentially a mixture of equivalent amounts of plain glucose 5% and acetated Ringers solution. Little is known about how fast glucose 2.5% can be infused during laparoscopic surgery without causing blunt hyperglycaemia, which results in osmotic diuresis when the baseline concentration is almost tripled7 and in a worse neurological outcome in case of cardiac arrest, which has been reported in laboratory animals having a 5-fold increase in plasma glucose.810 Furthermore, the volume effect of the fluid has not been studied during surgery, although rapid infusion of glucose 2.5% certainly affects the need for plasma volume expanders.
The aim of the present study was to describe the distribution and elimination of the glucose and volume components of glucose 2.5% during laparoscopic cholecystectomy. The glucose kinetics were analysed using traditional pharmacokinetic methods while the fluid was studied using a volume kinetic model which considered the osmotic translocation of water associated with the uptake of glucose into the cells.11 12 Nomograms were constructed to predict how fast glucose 2.5% needs to be infused to yield various predetermined glucose concentrations and dilutions of venous plasma.13 14
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The patients arrived at the hospital at 7 am and received no fluid or food before the surgery started at 8.30 am. All of them were given 2.57.5 mg of ketobemidone by i.m. injection for pre-medication about 1 h before entering the operating theatre. A cannula was placed in the cubital vein of each arm for the respective purposes of sampling blood and infusing fluid. The sampled plasma volume was replaced by the equivalent amount of saline 0.9%.
Without infusing any fluid, general anaesthesia was induced with thiopental 5 mg kg1 and maintained with intermittent doses of 0.1 mg of fentanyl every 30 min and continuous administration of a gas mixture containing sevoflurane 13% in oxygen and ambient air. Tracheal intubation was facilitated by i.v. injection of 0.5 mg kg1 of rocuronium. Monitoring of the patients included continuous pulse oximetry and electrocardiography.
When the first skin incison was made, an i.v. infusion of iso-osmotic glucose 2.5% with electrolytes (Na+ 70, Cl 45 and acetate 25 mmol litre1; Rehydrex, Pharmacia, Uppsala, Sweden) was started. As most patients receive about 1.5 litres of fluid in connection with cholecystectomy at our hospital, we aimed at administering this volume to an 80-kg patient over 60 min, which was done via an infusion pump (Flo-Gard 6201, Baxter Healthcare Ltd, Deerfield, IL, USA). After adjustment for body weight, the actual amount of fluid was 1413 (SD 193) ml, which was infused at a rate of 23.6 (3.2) ml min1. The dose of glucose was 35.3 (4.8) g or 196 (27) mmol. No more fluid was given during the 240-min experiment.
Measurements
Venous blood was collected every 5 min during the first 90 min and every 10 min for up to 240 min. The plasma glucose concentration was measured in single samples with the GLU Gluco-quant reagent (Roche Diagnostic Inc., Mannheim, Germany) on a Hitachi 917 (Hitachi Co., Naka, Japan). The blood haemoglobin (Hb) concentration, the red blood cell count (RBC) and the mean corpuscular volume (MCV) were measured in duplicate samples, and the baseline in triplicate samples, using Technicon Advia 120 (Bayer, Tarrytown, NY, USA). Venous blood was also withdrawn every 30 min for measurements of the serum sodium concentration by the Hitachi 917 system (Hitachi Co., Naka, Japan) and every 60 min for plasma insulin using an ELISA kit (Mercodia AB, Uppsala, Sweden). The coefficient of variation for the 800 duplicate samples was 1.3% for Hb, 1.2% for RBC, and 0.2% for MCV, while serial testing yielded 5% for insulin.
The heart rate and arterial pressure were measured using an automatic device (Datex-Ohmeda Instrumentation Inc., Box 900, Finland) after each blood sampling procedure.
All patients had an indwelling urinary catheter, which was used to obtain data on the urine volume and the concentrations of glucose and sodium and potassium as measured by the Hitachi 917 system.
Glucose kinetics
The kinetics of glucose were calculated according to principles described in two previous laboratory studies,11 12 but here the endogenous production of glucose (ki en) was also estimated. The plasma concentration (Cex) at any time (t) resulting from exogenous administration of glucose at the rate ki ex was calculated from the following equation:
in which Vd is the volume of distribution and CL the clearance of glucose. The glucose concentration (Cen) resulting from endogenous production of glucose (ki en) is given by:
The measured plasma glucose concentration is the sum of Cex and Cen. Naturally, the endogenous production is responsible for the baseline concentration, and Cen best reflects the production just before the infusion starts. Fitting these equations to the glucose data yielded the optimal estimates of Vd, CL, and ki en.
Elimination of glucose from Vd represents active transport of glucose into the cells. The uptake of glucose to the cells was calculated individually for each time increment by using Vd as the calibration factor between plasma glucose level and the amount of glucose in Vd. The precise equations used are given elsewhere.11
Volume kinetics
The distribution and elimination of infused fluid was studied by volume kinetic analysis. In the model used (Fig. 1), an i.v. infusion given at a constant rate (ki) enters a body fluid space having the volume (v1) which strives to be maintained at the baseline volume V1 by two elimination mechanisms; first, elimination occurred at a controlled rate proportional by a constant kr to the expansion of V1 and, secondly, at a fixed rate at 0.5 ml min1, which represents insensible fluid losses.14
|
Calculations
The model parameters for the glucose and volume components of glucose 2.5% were calculated in each individual patient using Matlab version 6.5.1 (Math Works Inc., Natick, MA, USA). The kinetic models used a non-linear least-squares regression routine based on a modified Gauss-Newton method. Input data obtained at baseline (time 0) and at any later time (t) comprised Hb, RBC and MCV and the baseline haematocrit. They were used to calculate the plasma dilution, which also represented the dilution of V1:
Corrections were made for iatrogenic dilution caused by the blood sampling, which averaged 7 ml for each sampling procedure. No correction for surgical bleeding was made; this amounted to 120 ml in one patient while no blood loss was recorded in the others. Weights inversely proportional to the predicted dilution plus 0.1 were applied. Output consisted in V1 and kr along with the uncertainty of these estimates, expressed as their SD.
Nomograms were constructed by simulating the expected plasma glucose and plasma dilution responses to various theoretical infusions of glucose 2.5% based on the numerical solutions to the differential equations describing the kinetic models.13 14 The best estimate of the model parameters (Table 1) were inserted into the solutions which had been programmed into the Matlab software.
|
The results are expressed as the mean (SD). Because of skewed distributions, the median (interquartile range) was chosen where appropriate. Differences were evaluated by analysis of variance (ANOVA). Correlations between parameters were studied by simple and multiple linear regression. P<0.05 was considered significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
Volume kinetics
The plasma dilution increased gradually during the infusion, too, and decreased to a quasi-steady state 30 min later (Fig. 4A). The dilutiontime curves showed more variability than the glucose concentration, although the kinetic analysis yielded an acceptable average curve during surgery (Fig. 4B).
|
The size of the expanded central body fluid space (V1) averaged 3.89 litres and the elimination rate constant (kr) was 36.7 ml min1 (Table 1, bottom).
Haemodynamic variables and urine
The systolic and diastolic arterial pressures decreased during induction of anaesthesia (P<0.001) but increased when surgery and the glucose infusion started (P<0.01). Thereafter, the arterial pressures and the heart rate remained fairly constant until the end of anaesthesia 85 min later (Fig. 5).
|
The total urinary excretion of sodium was 12.1 (7.118.3) mmol, which is 12.8 (6.418.2)% of the administered amount of sodium. Minimal glycosuria occurred in most patients, amounting to 0.4 (0.16.4) mmol, which should be compared with the average amount of infused glucose of 196 mmol.
Serum sodium decreased from 139.3 (2.2) mmol litre1 at baseline to 136.1 (1.9) mmol litre1 at 90 min (P<0.001) but the concentration was then restored partly. The sodium dilution method indicated an intracellular accumulation of 433 (151) ml of fluid at 120 min and 312 (222353) ml at 240 min (P<0.001 vs baseline).
Simulations
The plasma glucose concentration expected to result from infusing glucose 2.5% at various rates is given in the form of a nomogram in Figure 6. A corresponding nomogram for the expected plasma dilution is shown in Figure 7.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The volume of distribution (Vd) for the administered glucose molecules of 9.1 litres does not differ from that in the previous studies, but the clearance (CL) for glucose was only one-third of that obtained under laboratory conditions.11 12 The difference in the rate of elimination may also be illustrated by comparing half-lives, which is given by Vd * ln 2/CL. The half-life for glucose was 30 min during cholecystectomy while ranging between 11 and 16 min in the previous five series of glucose infusions in volunteers. The slow elimination of glucose made the patients prone to develop hyperglycaemia. Plasma glucose reached a peak of 16 mmol litres1, but this was followed by only minor glycosuria. Computer simulation indicates that the same infusion would increase plasma glucose to 10 mmol litre1 in a healthy volunteer. The low CL should prevent rebound hypoglycaemia, however, which is a potential problem when a brisk glucose infusion is stopped abruptly.
The low CL for glucose cannot be explained by insulin deficiency as the plasma concentrations of this hormone were very similar to the levels measured in the volunteer studies in which CL was three times higher.11 12 Therefore, insulin resistance was undoubtedly present15 although laparoscopic cholecystectomy is only a moderately stressful operation.5 Another study shows that glucose uptake was reduced by 20% the day after this operation6 while the present results were obtained during the surgical procedure. Regardless of trauma, however, insulin resistance develops when fasting is long enough to markedly reduce glycogen stores.16 The insulin resistance can be abolished by infusing large amounts of glucose before surgery17 or glucose with insulin during surgery.18 In the present study, a relative deficit in the glucose supply is indicated by the endogenous glucose production, which was low (1.1 mmol min1) compared with that in volunteers subjected to similar glucose infusions in whom we calculated the same rate in retrospect (mean 2.6 mmol min1).12 In both studies, the endogenous glucose production correlated closely with CL for glucose (Fig. 3A). This implies that the endogenous glucose production increased with larger amounts of glucose being transported into the cells. This correlation could be expected as CL varied much more than the baseline glucose level. All subjects were in the fasting state, however, which makes anaesthesia and surgical trauma the most apparent factors that differed between the patients and the volunteers. The patients were 10 yr older than the volunteers and mostly women, but this difference is less likely to account for the difference in CL as the body weights were the same in the groups (75 and 77 kg).
The dilutiontime curves showed a higher degree of inter-subject variability than for plasma glucose. A prompt plasma dilution occurred in response to the first fraction of the volume load, which might be explained by the vasodilatation resulting from the induction of anaesthesia, which allows more fluid to fill the vascular system.19 Although no fluid was infused, a plasma dilution of 5% was associated with the induction (data not shown), but this might not have fully accounted for the vasodilatation. Moreover, the surgical technique involves i.p. insufflation of carbon dioxide, which could have affected the early plasma dilution response to glucose 2.5% by modifying the cardiovascular regulatory centres.20
The awakening from anaesthesia disturbed the dilutiontime curve, which made it necessary for the kinetic analysis to involve the period of anaesthesia alone. As compared with laboratory conditions, the size of the body fluid space expanded by the infused fluid (V1) was 1020% larger while the elimination rate constant (kr) was only between one-half and one-third as high.11 12 This means that, in addition to the reduced distribution of fluid to V3 resulting from insulin resistance, the rate of elimination of fluid from the kinetic system was also decreased in a way similar to the elimination of Ringers solution during anaesthesia18 and trauma surgery.21 As kr corresponds to CL in conventional kinetics, the CL for glucose and fluid appeared to be reduced to the same degree during laparoscopy.
The decrease in the rate of elimination of fluid is even greater if we compare real data. Only 9% of the fluid infused was excreted as urine after 4 h while the figure was 62% under non-surgical conditions.12 Thus, the renal elimination of fluid was in fact reduced twice as much by the surgery as indicated by the kinetic analysis. To examine how much fluid is present in the kinetic system, one can multiply the size of V1 by the dilution at any time and estimate the excess volume in V3 in the same way. At 120 min, 175 ml of excess fluid was present in V1 and almost 500 ml in V3, the latter figure corresponding quite well to the intracellular accumulation of fluid of 443 ml estimated by the sodium dilution method. With a urinary excretion of only 48 ml at this time, the amount of missing fluid is as large as 700 ml. This discrepancy has not been encountered in our volunteer experiments.11 12
Excessive evaporation from the airways during general anaesthesia might account for part of this volume, but there are two other possible explanations for the missing fluid. The first is that the glucose solution occupied the fluid space between V1 and V3, which has been considered but not supported by the mathematical analysis. Computer simulations show that this space (V2) is likely to be expanded when CL and kr are very low. In this study, the shape of the dilutiontime curve typical for the presence of an intermediate body fluid space expanded by fluid was not present, but the relatively high variability of the curves and the fact that the curve-fitting was based on data from only 2 h make it impossible to rule out such a possibility. The second explanation relates to the finding that isoflurane anaesthesia creates a similar discrepancy between model-predicted elimination and renal excretion when saline 0.9% is infused in sheep.22 The effect is not a result of the mechanical ventilation but to the isoflurane itself.23 In the present study, anaesthesia was maintained with sevoflurane, which might cause the same third-spacing effect as isoflurane. Where the fluid resides is not known but, in any event, it does not readily equilibrate with V1.
The results of the kinetic analyses were illustrated by nomograms, which may be useful in clinical practice. They show that infusing glucose 2.5% at a constant rate throughout laparoscopic cholecystectomy is not optimal therapy. To prevent hyperglycaemia and progressive volume expansion, the infusion rate should be high at the beginning of surgery and then slowed down. In the nomogram for glucose, the anaesthetist can choose an acceptable increase in plasma glucose during and immediately after laparoscopy (Fig. 6). After having selected an acceptable target rise in plasma glucose in the left part of the nomogram, any combination of infusion rates or infusion times can be used to hit the isobar showing the accepted increase. Once this target is reached, the task is then to infuse enough glucose 2.5% to maintain the plasma concentration. This is done by following the isobar to the right and turning down the infusion rate to the value obtained in that part of the figure.
The nomogram for plasma dilution is based on the same principle as the one for plasma glucose, the only difference being that the steady-state level varied slightly, about 6% between a 10- and a 60-min infusion, because of variable amounts of fluid returning from V3 after metabolism of the glucose. The volume effect of glucose 2.5% is not negligible and should be considered when administering the fluid (Fig. 7). For example, an increase of plasma glucose by 2 mmol litre1 to slightly above the normal range can apparently be maintained by supplying glucose 2.5% at a rate of 3 ml min1 (180 ml h1). The infused fluid volume then dilutes the plasma by 7%, which corresponds to a plasma volume increase of about 250 ml in a normal-sized patient.
In conclusion, kinetic analysis of glucose 2.5% solution during laparoscopic cholecystectomy showed low values for endogenous glucose production and for the clearance of glucose and infused fluid volume. This results in an altered disposition of the fluid.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Cuschieri Sir A, Steele RJC, Moossa AR. Essential Surgical Practice, 4th Edn. London: ArnoldHodder Headline Group, 2001; 432
3 Sieber FE, Smith DS, Traystman RJ, Wollman H. Glucose: a reevaluation of its intraoperative use. Anesthesiology 1987; 67: 7281[ISI][Medline]
4 Miller RD. Anesthesia, 5th Edn. Philadelphia: Churchill Livingstone, 2000; 1596, 1606, 2012
5 Schauer PR, Sirinek KR. The laparoscopic approach reduces the endocrine response to elective cholecystectomy. Am Surg 1995; 61: 10611[ISI][Medline]
6 Thorell A, Nygren J, Essén P, et al. The metabolic response to cholecystectomy: insulin resistance after open compared with laparoscopic operation. Eur J Surg 1996; 162: 18791[ISI][Medline]
7 Doze VA, White PF. Effects of fluid therapy on serum glucose levels in fasted outpatients. Anesthesiology 1987; 66: 2236[ISI][Medline]
8 Myers RE, Yamaguchi S. Nervous system effects of cardiac arrest in monkeys. Arch Neurol 1977; 34: 6574[Abstract]
9 Siemkowicz E. Hyperglycemia in the reperfusion period hampers recovery from cerebral ischemia. Acta Neurol Scand 1981; 64: 20716[ISI][Medline]
10 de Courten-Myers GM, Yamaguchi S-I, Wagner KR, Ting P, Myers RE. Brain injury from marked hypoxia in cats: role of hypotension and hyperglycemia. Stroke 1985; 16: 101621[Abstract]
11 Sjöstrand F, Edsberg L, Hahn RG. Volume kinetics of glucose solutions given by intravenous infusion. Br J Anaesth 2001; 87: 83443
12 Sjöstrand F, Hahn RG. Validation of volume kinetic analysis of glucose 2.5% solution given by intravenous infusion. Br J Anaesth 2003; 90: 6007
13 Hahn RG, Svensén C. Plasma dilution and the rate of infusion of Ringers solution. Br J Anaesth 1997; 79: 647
14 Drobin D, Hahn RG. Volume kinetics of Ringers solution in hypovolemic volunteers. Anesthesiology 1999; 90: 8191[ISI][Medline]
15 Brandi LS, Frediani M, Oleggini M, et al. Insulin resistance after surgery: normalization by insulin treatment. Clin Sci 1990; 79: 44350[ISI][Medline]
16 Nygren J, Thorell A, Brismar K, Karpe F, Ljungqvist O. Short-term hypocaloric nutrition but not bed rest decrease insulin sensitivity and IGF-I bioavailability in healthy subjects: the importance of glucagon. Nutrition 1997; 13: 9455[CrossRef][ISI][Medline]
17 Ljungqvist O, Thorell A, Gutniak M, Häggmark T, Efendic S. Glucose infusion instead of preoperative fasting reduces postoperative insulin resistance. J Am Coll Surg 1994; 178: 32936[ISI][Medline]
18 Nygren JO, Thorell A, Soop M, et al. Perioperative insulin and glucose infusion maintains normal insulin sensitivity after surgery. Am J Physiol 1998; 275: E1408[ISI][Medline]
19 Ewaldsson C-A, Hahn RG. Volume kinetics during induction of spinal and general anaesthesia. Br J Anaesth 2001; 87: 40614
20 Odeberg S, Ljungqvist O, Svenberg T, Sollevi A. Lack of neurohumoral response to pneumoperitoneum for laparoscopic cholecystectomy. Surg Endosc 1998; 12: 121723[CrossRef][ISI][Medline]
21 Svensén C, Ponzer S, Hahn RG. Volume kinetics of Ringer solution after surgery for hip fracture. Can J Anaesth 1999; 46: 13341[Abstract]
22 Brauer KI, Svensén C, Hahn RG, Traber L, Prough DS. Volume kinetic analysis of the distribution of 0.9% saline in conscious versus isoflurane-anesthetized sheep. Anesthesiology 2002; 96: 4429[CrossRef][ISI][Medline]
23 Connolly CM, Kramer GC, Hahn RG, et al. Isoflurane but not mechanical ventilation promotes extravascular fluid accumulation during crystalloid volume loading. Anesthesiology 2003; 98: 67081[CrossRef][ISI][Medline]