Intravascular volume is a major determinant of venous return, and hence cardiac output. It is frequently altered during critical illness (1), being the target of therapeutic intervention in many intensive care unit (ICU) patients. Although the quantitative assessment of intravascular volume is highly desirable, this is not yet possible in routine clinical practice.
Ishihara et al. (2) have proposed a method of estimating circulating blood volume based on the calculation of the initial distribution volume of glucose (IDVG). The technique is simple, minimally invasive, and inexpensive. The estimation of IDVG relies on a standard one-compartment pharmacokinetic model of the initial rate of disappearance of an IV glucose bolus (3,4). IV glucose is rapidly distributed throughout the extracellular fluid compartment, including plasma volume and the interstitial fluid of highly perfused organs (5). It is unaffected by metabolism and insulin action (6). IDVG was studied in various groups of critical care situations, such as burns (7), esophageal surgery (8), myocardial infarction, and sepsis (9). In these studies, IDVG was correlated with cardiac output determined by thermodilution, and was described as a good indicator of preload. However, the individual responses to fluid therapy were not assessed.
We hypothesized that IDVG could be used to evaluate the plasma volume status of individual patients. Specifically, we evaluated the effect of a fluid challenge on IDVG and standard hemodynamic variables. We also studied the individual reproducibility of IDVG.
This investigation was approved by the ethics committee of our institution. After preoperative written, informed consent was obtained, patients undergoing elective cardiac surgery with cardiopulmonary bypass (CPB) were investigated in the ICU during the immediate postoperative period. Exclusion criteria were diabetes type 1 or 2, hypoglycemia <4 mmol/L, hyperglycemia >15 mmol/L, neurologic illness, and hemodynamic instability (requiring dobutamine infusion >10 μg · kg−1 · min−1, dopamine >10 μg · kg−1 · min−1, or norepinephrine >10 μg/min).
Patient management was performed by the attending ICU staff and data regarding IDVG were collected by a research nurse. Arterial catheter and a central venous catheter or a pulmonary artery thermodilution catheter were placed in the operating room. Patients were transferred tracheally intubated to the ICU, where management included continuous monitoring of arterial blood pressure, electrocardiograph, and Spo 2, blood transfusion to maintain hemoglobin concentration ≫8.0g/dL, and dobutamine and norepinephrine to maintain mean arterial blood pressure (MAP) >70 mm Hg. During the investigation, patients were tracheally intubated and mechanically ventilated, sedated with a continuous infusion of propofol and morphine. They received 60 mL/h of gluco-saline infusion. Respiratory management, sedation, and catecholamine infusion were not changed, and diuretics were not administered during the study period.
Upon ICU admission and after a clinical evaluation by the attending physician, patients were assigned to one of two groups, according to their fluid requirements. Patients requiring fluid therapy were assigned to the fluid group. IDVG was determined before and after a fluid challenge of 7 mL/kg of poly(O-2-hydroxyethyl) starch (Voluven; Fresenius, Bad Homburg, Germany) infused over 30 min. Patients not requiring fluid therapy were assigned to the control group, where IDVG was determined 3 times at 30-min intervals.
Central venous pressure (CVP) and MAP were recorded at each new measurement of IDVG. Blood loss through drainage was recorded. Because most low-risk hemodynamically stable patients did not require a pulmonary artery catheter, we did not consider cardiac output and wedge pressure.
For IDVG measurements, 5 g of glucose (50% glucose solution) was infused into the central venous catheter over 30 s (10). Serial blood samples were obtained through a radial or femoral artery catheter before and 3, 4, 5, 6, and 7 min after glucose injection. The catheter was flushed with a small amount of normal saline between measurements. Glucose plasma concentrations were determined with a Beckman glucose analyzer II (Beckman Instruments, Fullerton, CA). Each value was measured in duplicate and the average value was used.
The value of glucose at the initial time of distribution (C0) was obtained by back-extrapolation to time 0, using a one-compartment exponential model, and a least square fit on the decay of plasma glucose concentrations from time 3 to 7 min. The IDVG was calculated as the dose divided by the incremental glucose concentration (C0 minus glucose before injection), and was expressed in milliliter/kilogram body weight. A Microsoft Excel 2000 spreadsheet was used for these computations.
Statistical analysis was performed using JMP statistical software version 3.5.1 (SAS Institute, Cary, NC). P < 0.05 was considered statistically significant. The values of IDVG, regression coefficients, CVP, and MAP were compared by a two-way analysis of variance for the effect of group and time. Post hoc comparisons between groups at baseline, and over time within groups, were performed using the Scheffé and Dunnett’s tests. In the control group, the variation coefficients of the three measurements of IDVG were calculated for each patient as standard deviation/mean.
Twenty-six patients were enrolled. Two patients were excluded, one because of massive blood loss (fluid group) and one for a manipulation error (control group). Twenty-four patients (21 men and 3 women) completed the investigation, and were included in the analysis. Surgical procedures consisted of 17 coronary artery bypass graft, 5 valve, 1 combined operation, and 1 foramen ovale closure. All patients recovered uneventfully. Their clinical characteristics are described in Table 1. Their temperature increased from 35.9º ± 0.5ºC to 36.4 ± 0.7ºC in the fluid group, and from 35.9º ± 0.4ºC to 36.1º ± 0.5ºC during the second measurement, and to 36.6º ± 0.6ºC during the third measurement in the control group.
Basal glycemia was 8.4 ± 1.8 mmol/L. Figure 1 describes the increase and exponential decrease in glucose concentration in a representative patient. Glucose injection acutely increased glycemia by 3.1 ± 0.6 mmol/L. An exponential regression was obtained from all glucose dilutions, the regression coefficient (R2) ranged from 0.86 to 1, with a mean value of 0.96 (Table 2). There was no difference in R2 between groups.
Calculated baseline IDVG was 88.9 ± 16.9 mL/kg in the fluid challenge group, and 92.0 ± 27.4 in the control group. Mean IDVG increased by 5.7 ± 17.1 mL/kg after the fluid challenge, a change that was not significant. The variability of the response to fluid challenge was substantial (range from +47 to −41 mL/kg). The power of the comparison to detect a difference after fluid challenge was 0.25 for a P value of 0.05. CVP increased significantly during the fluid challenge, whereas there was no change in MAP (Fig. 2).
In the control group, there was no difference in IDVG, CVP, and MAP over time. The average coefficient of variation was 0.15 ± 0.08 (Fig. 3).
The main finding of the present study was that the IDVG method was insensitive in detecting the response to a standard fluid challenge during the phase immediately after cardiac surgery.
Several reasons may explain this negative finding. First, the reproducibility of IDVG must be considered. It had not previously been evaluated in the clinical setting (9,11). The control group of the present study allowed evaluation of IDVG reproducibility: We found an intraindividual variation coefficient of 15%, meaning that the method can reliably detect only a difference >30%. Thus, the use of IDVG to assess volemia in a clinical setting is obviously limited. Because we have no data in the literature to compare our results with others, we cannot exclude the possibility that the method might be more precise when used by others. However, we strictly followed the published description of the method (10), the procedure was performed by an experienced research nurse, and a precise timing for blood sampling and an accurate instrument for glucose measurement were used. If pre-analytical or analytical problems account for the limited precision of IDVG, it may remain impractical for clinical use.
One might argue that the fluid challenge was insufficient to produce a significant hemodynamic effect in our patients. This possibility seems unlikely, because 7 mL/kg is frequently used in critically ill patients. In a similar study (500 mL of hydroxyethyl starch over 30 min), a significant increase in the intrathoracic blood volume and stroke volume were observed using transpulmonary thermodilution (12). We observed a significant increase in CVP after infusion, suggesting that significant hemodynamic changes occurred during the fluid challenge. Otherwise, the distribution of glucose, which reflects the extracellular fluid compartment, differs from the distribution of polystarch, which mostly reflects the plasma. Thus, hydroxyethyl starch may have increased plasma volume only without contributing significantly to the extracellular fluid increase.
The present study has several limitations. It was not randomized, and fluid management was left to the clinical evaluation of the attending physician. Moreover, the patient population was limited to patients after cardiac surgery with CPB. These patients might have presented with hemorrhage, temperature change, alterations in vasomotor tone, or fluid shifts between compartments consecutive to CPB. Such phenomena might have modified plasma or extracellular fluid volumes, hence IDVG measurements. We cannot exclude that different results could be observed in other patient populations. The potential risk of increasing glycemia would likely preclude the use of such a method in patients with neurologic insult, thus limiting the categories of patients susceptible to be investigated with the IDVG method.
We conclude that, in a small group of ICU patients during the early postoperative period after cardiac surgery with CPB, the IDVG was not sensitive to a standard fluid challenge with a colloid solution. The most likely explanation is a weak precision of IDVG
The authors are deeply indebted to the nursing team of the surgical ICU for their active collaboration.
1.Peters J, Mack G, Lister G. The importance of the peripheral circulation in critical illnesses. Intensive Care Med 2001;27:1446–58.
2.Ishihara H, Shimodate Y, Koh H, et al. The initial distribution volume of glucose and cardiac output in the critically ill. Can J Anaesth 1993;40:28–31.
3.Suzuki A, Ishihara H, Okawa H, et al. Can initial distribution volume of glucose predict hypovolemic hypotension after radical surgery for esophageal cancer? Anesth Analg 2001;92:1146–51.
4.Matsui A, Ishihara H, Suzuki A, et al. Glucose dilution can detect fluid redistribution following phentolamine infusion in dogs. Intensive Care Med 2000;26:1131–8.
5.Iwakawa T, Ishihara H, Takamura K, et al. Measurements of extracellular fluid volume in highly perfused organs and lung water in hypo- and hypervolaemic dogs. Eur J Anaesthesiol 1998;15:414–21.
6.Steil GM, Richey J, Kim J, et al. Extracellular glucose distribution is not altered by insulin: analysis of plasma and interstitial L-glucose kinetics. Am J Physiol 1996;271:E855–64.
7.Ishihara H, Otomo N, Suzuki A, et al. Detection of capillary protein leakage by glucose and indocyanine green dilutions during the early post-burn period. Burns 1998;24:525–31.
8.Ishihara H, Matsui A, Muraoka M, et al. Detection of capillary protein leakage by indocyanine green and glucose dilutions in septic patients. Crit Care Med 2000;28:620–6.
9.Ishihara H, Iwakawa T, Hasegawa T, et al. Does indocyanine green accurately measure plasma volume independently of its disappearance rate from plasma in critically ill patients? Intensive Care Med 1999;25:1252–8.
10.Shimodate Y, Koh H, Ishihara H, Matsuki A. Comparison of glucose and sucrose as an indicator for dilution volumetry in haemorrhagic shock. Eur J Anaesthesiol 1995;12:397–401.
11.Ishihara H, Suzuki A, Okawa H, et al. The initial distribution volume of glucose rather than indocyanine green derived plasma volume is correlated with cardiac output following major surgery. Intensive Care Med 2000;26:1441–8.
© 2005 International Anesthesia Research Society
12.Michard F, Alaya S, Zarka V, et al. Global end-diastolic volume as an indicator of cardiac preload in patients with septic shock. Chest 2003;124:1900–8.