Methods for the quantitative assessment of electrolyte disturbances in hyperglycaemia

While empirical calculations are presently used, exact solutions to compute volume and solute changes of hyperosmolar coma (HC) can be obtained by subdividing the patients according to well defined clinical and laboratory conditions. These are represented by PNa(G), the plasma Na concentration that would be present if there were only glucose addition (GA), that discloses prevalent Na depletion when >PNa(1), prevalent water deficit when , = or

We built a mathematical model describing fluid derangements used to execute computer-simulated experiments of HC. The derangements were generated on the computer by adding, to the extra-cellular volume, different amounts of glucose while subtracting variable combinations of ions and solvent. The model yielded true solute concentrations from which our formulas computed the amounts lost or gained. These were identical to the true changes introduced to simulate the derangements (R(2)=1.00, P<0.0001) when the boundary conditions for PNa(G), exclusive NaCl loss and Posm(1)-Posm(0) were met. In patients with HC in whom these same boundary conditions were satisfied, the computations of glucose and Na changes with our new formulas were not significantly different from those estimated after correction of the derangements, considered true values (R(2)=0.60, P<0.05), and showed a satisfactory agreement with the clinical evaluation.

Our new methods are more accurate than the traditional ones, as they reach a better quantitative assessment of the entity of the derangements, avoiding iatrogenic dysnatraemias after correction of HC.