What are the differences between the Boston approach, Copenhagen approach, and Stewart approach in interpreting acid-base balance?

Acid-Base Balance Interpretation: Three Major Approaches

The three primary approaches to acid-base interpretation are: (1) the Traditional/Boston/Physiological approach (bicarbonate-centered), (2) the Copenhagen/Base Excess approach, and (3) the Stewart/Physicochemical approach—each offering distinct frameworks for understanding acid-base disorders, with the Traditional approach remaining most widely used in clinical practice while the Stewart approach is increasingly favored in critical care settings. 1, 2

The Traditional (Boston/Physiological) Approach

Core Principles

• This bicarbonate-centered method uses the Henderson-Hasselbalch equation as its foundation, focusing on pH, PaCO2, and HCO3- as the primary variables for assessment. 1, 2
• The approach evaluates pH (normal 7.35-7.45), PaCO2 (normal 35-45 mmHg), and HCO3- (normal 22-26 mEq/L) to identify primary disorders and compensatory responses. 3
• It classifies disorders into four primary categories: respiratory acidosis (pH < 7.35, PaCO2 > 45 mmHg), respiratory alkalosis (pH > 7.45, PaCO2 < 35 mmHg), metabolic acidosis (pH < 7.35, HCO3- < 22 mEq/L), and metabolic alkalosis (pH > 7.45, HCO3- > 26 mEq/L). 3

Clinical Application Algorithm

• Step 1: Assess pH to determine if acidemia or alkalemia is present. 4
• Step 2: Determine the primary disorder by identifying whether PaCO2 or HCO3- moves in the same direction as pH. 4
• Step 3: Calculate the anion gap ([Na+] - [Cl-] - [HCO3-]) to further classify metabolic acidosis as high or normal anion gap. 4
• Step 4: Assess for appropriate compensation using established formulas to determine if a mixed disorder exists. 4

Strengths and Limitations

• This approach works exceptionally well clinically and is recommended whenever serum total protein, albumin, and phosphate concentrations are normal. 1
• It provides a simpler bedside methodology with physiological evaluation and easily understandable severity assessment. 4
• The major limitation is that it treats HCO3- as an independent variable when it is actually dependent on other factors, which can lead to incomplete understanding in complex cases. 2, 5

The Copenhagen (Base Excess) Approach

Core Principles

• This method centers on base excess (BE) or base deficit as the primary metric for assessing metabolic acid-base disturbances, representing the amount of acid or base needed to return pH to 7.40 at PaCO2 of 40 mmHg. 1, 2
• Base excess is calculated using the Siggaard-Anderson nomogram or modern blood gas analyzers, with normal range being -2 to +2 mEq/L. 6
• The approach separates respiratory (PaCO2-driven) from metabolic (base excess-driven) components more distinctly than the Traditional approach. 2

Clinical Application

• Negative base excess (base deficit) indicates metabolic acidosis, while positive base excess indicates metabolic alkalosis, providing a single number to quantify metabolic derangement. 2
• This approach allows for assessment of contributions from different sources: free water changes (assessed by [Na+]), chloride changes, protein concentration changes, and "other species" (unmeasured ions). 6
• The base-excess gap (difference between measured base excess and that predicted from known contributors) identifies unmeasured anions or cations. 6

Strengths and Limitations

• The Copenhagen approach provides a more quantitative assessment of metabolic disturbances and is particularly useful for trending changes over time. 2
• It shares similar limitations with the Traditional approach in that it doesn't fully explain the mechanistic causes of acid-base disorders. 1
• The base-excess gap correlates with the anion gap and provides readily available estimates of unmeasured ions at the bedside. 6

The Stewart (Physicochemical) Approach

Core Principles

• This modern approach identifies three independent variables that determine acid-base status: (1) PaCO2, (2) Strong Ion Difference (SID), and (3) total weak acid concentration (primarily albumin and phosphate). 1, 5
• In this model, pH and HCO3- are dependent variables—they represent effects rather than causes of acid-base derangements and cannot be regulated directly. 5
• The Strong Ion Difference is calculated as the difference between strong cations (primarily Na+ and K+) and strong anions (primarily Cl-), with normal SID being approximately 40-42 mEq/L. 6, 5

Clinical Application Algorithm

• Step 1: Calculate the apparent SID from measured electrolytes: SID = [Na+] + [K+] - [Cl-] - [lactate]. 6
• Step 2: Calculate the effective SID from the Henderson-Hasselbalch equation using measured pH, PaCO2, albumin, and phosphate. 6
• Step 3: Determine the SID gap (difference between apparent and effective SID) to identify unmeasured ions. 6
• Step 4: Assess each independent variable's contribution: PaCO2 for respiratory component, SID for metabolic component (particularly sodium-chloride difference), and weak acids (albumin/phosphate) for their effects. 5

Mechanistic Insights

• Hyperchloremic metabolic acidosis occurs when chloride increases relative to sodium, decreasing the SID and thereby lowering pH—this explains why normal saline administration causes acidosis. 7
• The Stewart approach reveals that sodium bicarbonate works by increasing SID through raising plasma sodium, not by directly adding bicarbonate. 5
• Changes in albumin concentration significantly affect acid-base status: hypoalbuminemia causes alkalosis by reducing weak acid concentration. 5

Strengths and Limitations

• This approach is increasingly used by anesthesiologists and intensive care specialists and is specifically recommended when serum total protein, albumin, or phosphate concentrations are markedly abnormal, as commonly occurs in critically ill patients. 1
• It provides superior mechanistic understanding and explains phenomena that the Traditional approach cannot, such as dilutional acidosis and the effects of albumin changes. 2, 5
• The major limitation is complexity—it requires more calculations and is less intuitive for bedside use, though it has been largely ignored by general physiologists. 1
• Recent studies raise concerns about poor agreement between central laboratory and point-of-care measurements of plasma sodium and chloride, which are critical for Stewart calculations. 5

Comparative Analysis and Clinical Recommendations

When to Use Each Approach

• Use the Traditional approach for routine clinical scenarios with normal protein and electrolyte concentrations—it remains the most practical for general medical practice. 1, 4
• Deploy the Stewart approach in critical care settings, particularly when managing patients with abnormal albumin, massive fluid resuscitation, or unexplained acid-base disorders despite normal anion gap. 1, 5
• The Copenhagen approach serves as a middle ground, providing quantitative assessment while maintaining relative simplicity. 2

Integration of Approaches

• Although conceptually different, these approaches are complementary and provide, in principle, the same information about acid-base status when properly applied. 1
• The anion gap from the Traditional approach, base-excess gap from the Copenhagen approach, and SID gap from the Stewart approach all reflect unmeasured ions and correlate with each other. 6
• Clinicians can use the Traditional approach for initial assessment and switch to Stewart methodology when encountering complex cases with protein/electrolyte abnormalities. 2

Critical Clinical Pitfalls

• Arterial blood gas analysis alone may be misleading—even simultaneous arterial and central venous sampling may poorly estimate myocardial and cerebral intracellular acid-base status. 8
• Venous samples can acceptably indicate pH and CO2 in many situations, avoiding unnecessary arterial punctures. 8
• The anion gap should always be corrected for albumin changes, as hypoalbuminemia artificially lowers the anion gap and can mask high anion gap metabolic acidosis. 5
• When using the Stewart approach, ensure laboratory measurements are accurate—point-of-care devices may have significant discrepancies with central laboratory values for sodium and chloride. 5