Anesth Analg 2004;98:271-272
© 2004 International Anesthesia Research Society
LETTERS TO THE EDITOR
Stewart Approach Is Not Always a Practical Clinical Tool
Richard M. Effros, MD
Respiratory and Critical Care Medicine, Medical College of Wisconsin, Milwaukee, WI
To the Editor:
Dr. Constable presents a very lucid explanation of the strong ion concept in his recent editorial (1) and its application to understanding hyperchloremic acidosis. However, I would differ with his conclusion that the conventional Henderson-Hasselbalch approach cannot be used to understand this disorder. The decrease in pH that accompanies infusions of saline can be explained because the dilution in bicarbonate is not usually matched by a proportionate fall in PCO2, which is regulated by the respiratory center (2). The Stewart approach is elegant, and we have found it very useful under some circumstances. However, it is uncertain whether it will become a practical clinical tool. Data about strong ions, phosphate, and protein that are used with the strong ion equations to calculate pH and bicarbonate are frequently unavailable and must be estimated. In practice, the pH, PCO2 and bicarbonate are measured with great accuracy, and these values can be used with additional information concerning the conventional anion gap to diagnose most acid-base disorders. The reliability of HCO3- measurements is routinely checked by comparing values calculated from arterial pH and PCO2 and the venous CO2 content. It remains to be seen whether the "revolution" predicted by Dr. Constable is at hand.
References
- Constable PD. Hyperchloremic acidosis: the classic example of strong ion acidosis. Anesth Analg 2003; 96: 919–22.[Free Full Text]
- Garella S, Chang B, Kahn SI. Dilution acidosis and contraction alkalosis: a review of a concept. Kidney Intern 1975; 8: 279–83.[Web of Science][Medline]
Response
Peter D. Constable, BVSc, PhD
Department of Veterinary Clinical Medicine, College of Veterinary Medicine, University of Illinois at Urbana-Champaign, Urbana, IL
In Response:
Dr. Effros revisits the mechanism of dilutional acidosis as viewed from the Henderson-Hasselbalch approach to acid-base balance, an issue that has been discussed sporadically over the decades (1–4). The traditional Henderson-Hasselbalch explanation (which I will term the bicarbonatecentric view of acid-base equilibrium) states that infusion of large volumes of crystalloid solutions such as 0.9% NaCl decreases the plasma bicarbonate concentration without changing PCO2, thereby creating a metabolic acidosis. This purported mechanism, if true, assumes that the bicarbonate concentration is an independent determinant of plasma pH. In contrast, because the strong ion approach assumes that bicarbonate is a dependent variable that cannot, by itself, influence pH, the strong ion approach provides an alternative explanation for dilutional acidosis.
The IV administration of a crystalloid solution will alter two of the three independent determinants of plasma pH, namely strong ion difference (SID) and the total plasma concentration of nonvolatile buffers (Atot) (5). The effect on Atot is a direct and predictable effect of the volume infused; the greater the volume administered, the larger the decrease in Atot. Because a decrease in Atot causes a nonvolatile buffer ion alkalosis, the effect of volume infusion on Atot does not provide an explanation for dilutional acidosis. Instead, dilutional acidosis is primarily due to the effect of infusion on plasma SID; the net effect on plasma SID (and therefore pH) depends on the volume infused and the SID of the infused crystalloid solution.
The effect of IV crystalloid solution formulation on plasma pH is best depicted by plotting plasma pH against the volume infused for solutions of differing SID (Fig. 1). Using the 6 factor simplified strong ion equation (5) and recently determined values for Atot (0.224[total protein in g/L]) and Ka (0.8 x 10-7) of human plasma (6), while assuming normal values for plasma pH (7.40), PCO2 (40 mm Hg), pK1' (6.120), S (0.0307 [mM/L]/mm Hg), total protein concentration (7.0 g/dL), and SID (34 mEq/L), we can calculate the effect of infusing different crystalloid solutions on plasma pH. Four solutions were examined: 0.9% NaCl (SID = 0 mEq/L), lactated Ringer’s solution (theoretical maximum SID = 28 mEq/L), acetated Ringer’s solution (theoretical maximum SID = 50 mEq/L), and 1.3% NaHCO3 (SID = 155 mEq/L); the theoretical maximum SID represents the final SID after complete metabolism of metabolizable strong anions (lactate in lactated Ringer’s, acetate and gluconate in acetated Ringer’s). Figure 1 tells a clear story: 0.9% NaCl decreases plasma pH, lactated Ringer’s solution slightly increases pH, acetated Ringer’s solution is more strongly alkalinizing, and 1.3% sodium bicarbonate causes a marked increase in pH. In fact, based on normal values for human plasma, the IV administration of a crystalloid solution with an effective SID <23.4 mEq/L will always be acidifying, whereas the administration of a crystalloid solution with an effective SID >23.4 mEq/L will always be alkalinizing (Fig. 1).
View larger version (32K):
[in this window]
[in a new window]
|
Figure 1. Effect of rapid IV administration of different crystalloid solutions on plasma pH in humans. SID = strong ion difference, which is the difference between the charge of strong cations and strong anions in plasma.
|
|
Adherents to the bicarbonatecentric viewpoint will say there is nothing new here, because application of the Henderson-Hasselbalch equation to the normal plasma values for pH (7.40), PCO2 (40 mm Hg), pK1' (6.120), and S (0.0307 [mM/L]/mm Hg) produces a calculated bicarbonate concentration of 23.4 mM/L. Accordingly, the bicarbonatecentric argument is that infusion of a solution with an effective bicarbonate concentration >23.4 mM/L will increase plasma bicarbonate concentration and create a metabolic alkalosis, whereas infusion of a solution with a bicarbonate concentration <23.4 mM/L (such as 0.9% NaCl, where bicarbonate concentration = 0 mM/L) will decrease plasma bicarbonate concentration and thereby create a metabolic acidosis (Fig. 1). However, the bicarbonatecentric approach ignores the effect of fluid administration on decreasing Atot; for this reason the plasma pH-volume infused relationship for 0.9% NaCl calculated from the Henderson-Hasselbalch equation is always more acidic than that calculated from the strong ion approach (Fig. 1). In other words, an experimental study examining the effect of rapid IV administration of large volumes of 0.9% NaCl on plasma pH may be able to determine which one of the strong ion or bicarbonatecentric approaches provides a better description of the data.
The strong ion approach leads to the interesting conclusion that the rapid IV administration of equal volumes of 0.9% NaCl, 5% dextrose, Ringer’s solution, and mannitol will produce the same acidifying effect, because all have a SID = 0 mEq/L. Current application of strong ion difference theory therefore explains the experimental results of Asano et al. (7), who observed an identical decrease in plasma pH in dogs administered 0.9% NaCl, 5% dextrose, or 5% mannitol IV at 88 mL/kg body weight. Figure 1 provides an explanation as to why the rapid IV administration of large volumes (70 mL/kg) of 0.9% NaCl produces a metabolic acidosis in humans, whereas administration of equivalent volumes of lactated Ringer’s solution does not alter plasma pH (8). No doubt one of the reasons for the continuing popularity of lactated Ringer’s solution in fluid therapy is its minimal effect on plasma pH (Fig. 1). In fact, because lactated Ringer’s is a racemic mixture of L- and D-lactate, and because it is unlikely that all of the D-lactate is metabolized, the effective SID of lactated Ringer’s solution is <28 mEq/L, and may approximate 23.4 mEq/L.
Dr. Effros correctly raises the question as to whether an acid-base revolution is at hand. For the proponents of the bicarbonatecentric view of acid-base equilibria, the thoughts of Thomas Huhn on scientific revolutions (9) may provide an interesting perspective.
References
- Peters JP, van Slyke DD. Quantitative clinical chemistry: interpretations.Vol 1. Baltimore, MD: Williams & Wilkins, 1931: 971–2.
- Shires GT, Holman J. Dilutional acidosis. Ann Intern Med 1948; 28: 557–9.[Abstract/Free Full Text]
- Prough DS, Bidani A. Hyperchloremic metabolic acidosis is a predictable consequence of intraoperative infusion of 0.9% saline. Anesthesiology 1999; 90: 1243–5.[Web of Science][Medline]
- Hamill-Ruth RJ. Dilutional acidosis: a matter of perspective. Crit Care Med 1999; 27: 2296–7.[Web of Science][Medline]
- Constable PD. A simplified strong ion model for acid-base equilibria: application to horse plasma. J Appl Physiol 1997; 83: 297–311.[Abstract/Free Full Text]
- Staempfli HR, Constable PD. Experimental determination of net protein charge and Atot and Ka of nonvolatile buffers in human plasma. J Appl Physiol 2003; 95: 620–30.[Abstract/Free Full Text]
- Asano S, Kato E, Yamauchi M, et al. The mechanism of acidosis caused by infusion of saline solution. Lancet 1966; 1: 1245–6.[Web of Science][Medline]
- Scheingraber S, Rehm M, Sehmisch C, Finsterer U. Rapid saline infusion produces hyperchloremic acidosis in patients undergoing gynecologic surgery. Anesthesiology 1999; 90: 1243–54.
- Kuhn TS. The structure of scientific revolutions. 2nd ed. Chicago IL: University of Chicago Press, 1970.
This article has been cited by other articles:
|
|
|
|
 
I. Kurtz, J. Kraut, V. Ornekian, and M. K. Nguyen
Acid-base analysis: a critique of the Stewart and bicarbonate-centered approaches
Am J Physiol Renal Physiol,
May 1, 2008;
294(5):
F1009 - F1031.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
