Facts for estimation:

(1) Most of the total body potassium is intracellular.

(2) Most of the total body sodium is extracellular.

(3) The ratio of sodium and potassium to water is a constant for an individual and all her or his tissues. This ratio is representative for a subset like peripheral blood.

Method:

(1) Measure total body water in liters. This can be done using isotope dilution with deuterium, tritium or oxygen-18.

(2) Measure total body sodium using isotope dilution with sodium-22.

(3) Calculate the R value for peripheral whole blood, where:

R =

= ((concentration of sodium in mEq/L for whole blood) + (concentration of potassium in mEq/L for whole blood)) / (water in liters per liter whole blood)

(4) Estimate the total exchangeable potassium as follows:

total exchangeable potassium in mEq =

= ((R) * (total body water in liters)) – (total exchangeable sodium in mEq)

NOTE: I was unable to find a specific procedure for calculating R from the peripheral blood. Based on how I read the description by Shizgal (page 1186-1188), I believe these are the steps involved. BEFORE USING IN A CLINICAL SETTING, THIS NEEDS TO BE VERIFIED:

(1) A 1 mL sample of peripheral blood is initially weighed then dried down to a constant weight.

(2) The difference in weights is the weight of water present. Since the density of water is 1, this weight is also the volume of water present in 1 mL blood.

(3) The volume of water in liters per liter of whole blood can then be calculated.

(4) Measurement of potassium and sodium are in mEq per liter whole blood. Shizgal says that it is necessary to "correct for water content". The goal here appears to be to convert mEq per liter whole blood to mEq per liter body water.

(5) The units should work out to be mEq per liter water. This appears to be the correct units for Equation 3 on page 1186 and Figure 1 on page 1187 (Shizgal).