INTRODUCTION: We previously reported that initial distribution volume of glucose (IDVG) reflects central extracellular fluid volume, and that IDVG may represent an indirect measure of cardiac preload that is independent of the plasma glucose values present before glucose injection or infusion of insulin and/or vasoactive drugs. The original IDVG measurement requires an accurate glucose analyzer and repeated arterial blood sampling over a period of 7 min after glucose injection. The purpose of the present study was to compare approximated IDVG, derived from just two blood samples, versus original IDVG, and to test whether approximated IDVG is an acceptable alternative measure of IDVG in the intensive care unit. METHODS: A total of 50 consecutive intensive care unit patients were included, and the first IDVG determination in each patient was analyzed. Glucose (5 g) was injected through the central venous line to calculate IDVG. Original IDVG was calculated using a one-compartment model from serial incremental arterial plasma glucose concentrations above preinjection using a reference glucose analyzer. Approximated IDVG was calculated from glucose concentrations in both plasma and whole blood, using a combined blood gas and glucose analyzer, drawn at two time points: immediately before glucose injection and 3 min after injection. Subsequently, each approximated IDVG was calculated using a formula we proposed previously. RESULTS: The difference (mean +/- standard deviation) between approximated IDVG calculated from plasma samples and original IDVG was -0.05 +/- 0.54 l, and the difference between approximated IDVG calculated from whole blood samples and original IDVG was -0.04 +/- 0.61 l. There was a linear correlation between approximated and original IDVG (r2 = 0.92 for plasma samples, and r2 = 0.89 for whole blood samples). CONCLUSION: Our findings demonstrate that there was good correlation between each approximated IDVG and original IDVG, although the two measures are not interchangeable. This suggests that approximated IDVG is clinically acceptable as an alternative calculation of IDVG, although approximated and original IDVGs are not equivalent; plasma rather than whole blood measurements are preferable.