Independence of Variables in Stewart’s Model of the Acid-Base Chemistry of the Blood Plasma

Abstract

Several approaches have been taken to modelling of the acid-base chemistry of blood. The Stewart approach includes three independent variables (SID, PCO2, Atot), which are postulated to completely describe causal mechanisms behind changes in acid-base status. This paper explores this postulate, simulating typical clinical examples using an online modeling tool. For changes in alveolar ventilation, production of strong acid, and selective removal of non-charged protein buffers, this postulate is true. However for non-selective protein buffer removal SID and Atot cannot be seen as independent. The paper discusses the implication of this on diagnosis of acid-base disturbances in patients with abnormal protein concentration.