On acid-base effects of albumin

Dear Editor,

We read with interest the recent review in this journal by Christian Wiedermann¹. We congratulate the author for producing a comprehensive review on the physiology of albumin and its role in perioperative and critically ill patients. However, we disagree with the statement “Albumin has a variety of biological properties and functions, including…buffering capabilities, which may be especially relevant in critically ill patients”¹.

Buffers were defined by van Slyke as “substances which by their presence in solution increase the amount of acid or alkali that must be added to cause a unit change in pH”². Plasma proteins, with albumin being the most abundant one, are regarded as effective buffers by modern textbooks of physiology³. However, exactly the opposite can be inferred from Stewart’s physical–chemical theory: increasing the concentration of albumin (the weak non-volatile acid by Stewart’s nomenclature) should in fact reduce the buffering ability of plasma^4,5.

We aimed to test in a direct experiment how albumin concentration in artificial plasma and blood affects the resistance to pH change in response to experimental pCO₂ manipulations. By using water, sodium chloride, hydrochloric acid, sodium bicarbonate and human albumin (Sigma-Aldrich A9511, purity ≥97%), we prepared five solutions of artificial plasma with varying concentrations of albumin (10, 30, 50, 70 and 90 g/L) and constant strong ion difference (SID, 39–41 mEq/L). Using a CO₂ tonometer (RNA Medical, Devens, USA) and gas mixtures containing 0–20% carbon dioxide and 21% oxygen, we manipulated pCO₂ of the solutions in a range of 15–120 mmHg (2–16 kPa) at 37 °C. Immediately after gas equilibration, pH and pCO₂ were measured using an ABL90 blood gas analyzer (Radiometer, Copenhagen, Denmark). At least 15 measurements were performed on each sample. For each albumin concentration, we calculated buffer power as the slope of the [H⁺]/pCO₂ regression line. Additionally, to increase biological plausibility of the model, we repeated the experiment with artificial blood created by adding red blood cells from a single healthy donor into the solutions of artificial plasma. The results are summarized in Figure 1. Increasing the concentration of albumin caused an upward shift of the [H⁺]/pCO₂ regression line and increased its slope. This means that more albumin makes the solution of artificial plasma more acidic, but its resilience against pH shifts in response to changing pCO₂ (“the buffer power”) is decreased. The same behavior was observed in solutions containing red blood cells (data not shown). Our simple experiments thus confirm, for the first time to our knowledge, Stewart’s theoretical predictions that “presence of the weak acid in plasma decreases, rather than augments, the buffering ability of plasma”⁴.

Figure 1. The relation between [H⁺] and pCO₂ in tonometered solutions of artificial plasma with varying concentration of albumin. Albumin (Alb) in g/L, SID in mEq/L and osmolarity (osmol) in mOsm/L.

In clinical practice it should be noted that by infusing a commercial albumin preparation, a clinician influences acid base status by: (1) increasing preload and hence tissue perfusion; (2) changing plasma SID by infusing the albumin solvent containing preparation-specific concentrations of strong ions and stabilizers, such as sodium octanoate and N-acetyltryptophan⁶; and (3) increasing the albumin concentration in plasma which, apart from its predictable acidifying effect, leads to a reduction of buffering capacity of plasma. This counterintuitive effect is in accordance with predictions based on Stewart’s physical–chemical model and has been validated by our experiments.

Transparency

Declaration of funding

There is no funding to declare.

Declaration of financial/other relationships

No potential conflict of interest was reported by the authors.

Acknowledgements

This work was supported by Charles University, project GA UK No. 1004120 and we thank Thomas Langer for inspiration and help.