The determination of protein-bound calcium and magnesium by ultrafiltration- in-vivo: with special reference to terminology and method

Abstract

A modification of the ultrafiltration- in-vivo method introduced by Gerbrandy is presented by which the accuracy is increased sufficiently to allow it to be used for the determination of protein-bound cation in the individual patient. First the term “protein-bound” is discussed, and it is argued that for a correct analysis of cation-binding in clinical conditions the protein-bound fraction should be given in mequiv. bound per unit protein, together with the protein composition and the concomitant concentration of free cation. As to the method, the increase in accuracy is brought about by prolonging the period of venous compression of the arm to at least 25 min and by increasing the number of blood samples taken. Provided that the arm is kept completely relaxed, venous compression at a pressure of 90–100 mm Hg is well tolerated and can be maintained for up to 35 min without the occurrence of lactate formation or a serious fall in venous pH: in a total of 82 experiments the mean decrease in the latter was 0.05 and the maximum decrease 0.09. Although the mean increase in total protein was considerable, i.e. 56% of the precompression concentration, no change in protein composition occurred. With the additional aid of equilibrium dialysis it was found that at constant free ( i.e. filterable) cation concentration, a linear relation existed between total concentration of Ca, Mg and Na, respectively and of protein concentration over the range 0–130 g/kg H 2O. Free and protein-bound ( i.e. non-filterable) cation are therefore determined by calculating regression lines from the cation-protein relation found in each experiment. In 15 healthy men the mean values for free and protein-bound Ca so obtained were respectively 2.92 mequiv./kg H 2O and 2.98 mequiv. per 100 g of total protein. For Mg the values 1.13 mequiv./kg H 2O and 0.89 mequiv. per 100 g of total protein were found. The agreement with results obtained by others with different methods was good, and in the case of Ca the spread was comparable with that published for in-vvvitro-ultrafiltration. Observations in patients with various diseases among which were abnormal protein composition and hypercholesteremia gave further indication that in the case of Ca the method is sufficiently accurate to be clinically useful.