A structured approach to acid‐base interpretation

Understanding how to identify acid‐base disorders is essential, as their consequences can be life‐threatening

Disorders affecting acid‐base status are common, with a variety of pathophysiological causes and consequences that range from minor disturbances to life‐threatening disorders. A thorough understanding of how to identify acid‐base disorders (by history, examination and targeted investigations including blood gas) is essential. Through a series of clinical vignettes, this article illustrates the value of a structured approach to acid‐base assessment.

Acidaemia and alkalaemia are terms that define overall changes in pH (below or above normal range, respectively), while the pathophysiological processes that change the hydrogen ion (H⁺) concentration are respiratory or metabolic acidosis (increased H⁺) and respiratory or metabolic alkalosis (decreased H⁺).1 The Henderson–Hasselbach approach to pH homeostasis (used in this article) considers arterial partial pressure of carbon dioxide (Paco₂) and plasma concentration of bicarbonate (${\text{HCO}}_{{\mathrm{₃}}^{-}}$) the two fundamental variables. Changes in one of these variables cause predictable compensatory changes in the other variable to control pH — Box 1 depicts the methods of calculating expected compensation.3 A stepwise physiological approach, such as that presented in this article, is considered the simplest and most rigorous approach2 to assess acid‐base disorders.

The decision to sample arterial blood gases (ABGs) or venous blood gases (VBGs) depends on the individual clinical situation. Each has advantages and disadvantages (Box 2). A recent meta‐analysis4 suggests there is correlation in pH between ABGs and VBGs, but not between venous and arterial partial pressure of carbon dioxide (Pco₂) and partial pressure of oxygen (Po₂)_. Lactate and ${\text{HCO}}_{{\mathrm{₃}}^{-}}$ correlate well only when these values are normal or close to normal. ABG testing remains the gold standard in assessing acid‐base and respiratory status. For this reason, all the examples used in this article represent ABG analyses.

Vignette 1

A 70‐year‐old man with a history of chronic obstructive pulmonary disease (COPD) presents with dyspnoea. An ABG sample was analysed to assess oxygenation and ventilation: pH 7.30 (reference interval [RI], 7.35–7.45), Paco₂ 65 mmHg (RI, 35–45 mmHg), HCO₃− 34 mmol/L (RI, 22–26 mmol/L).

There is acidaemia (low pH), respiratory acidosis (high Paco₂) and metabolic alkalosis (high HCO₃−). The high Paco₂ level correlates with the low pH, so the primary or predominant problem is respiratory acidosis.

In acute respiratory acidosis, the HCO₃− should rise by 1 mmol/L for every 10 mmHg increase in Paco₂ (this patient's Paco₂ has risen by 25 mmHg), giving an expected HCO₃− of 24 + (2.5 × 1) = 26.5 mmol/L.

In chronic respiratory acidosis, the HCO₃− should rise by 4 mmol/L for every 10 mmHg increase in Paco₂, giving an expected HCO₃− of 24 + (2.5 × 4) = 34 mmol/L.

Discussion

Based on the patient's past medical history and the ABG, this patient has a chronic respiratory acidosis with appropriate metabolic compensation.

Respiratory acidosis may be caused by high carbon dioxide (CO₂) production or, more commonly, by poor CO₂ clearance. The minute ventilation (minute ventilation = tidal volume × respiratory rate) directly controls CO₂ clearance. Hence, pathologies that decrease the tidal volume and/or the respiratory rate cause respiratory acidosis (Box 3). This patient's chronic respiratory acidosis is likely to be caused by his known COPD — although other causes may need to be excluded.

Vignette 2

A patient presents 10 days after a long‐haul flight with pleuritic chest pain and dizziness. ABG was analysed to assess oxygenation and ventilation: pH 7.53, Paco₂ 20 mmHg, HCO₃−  19 mmol/L. There is alkalaemia (high pH) with low Paco₂ (respiratory alkalosis) and low HCO₃− (metabolic acidosis). The low Paco₂ level correlates with the high pH, indicating that respiratory alkalosis is the primary problem.

In acute respiratory alkalosis, for every 10 mmHg decrease in Paco₂, the HCO₃− should decrease by 2 mmol/L. The Paco₂ level has decreased by 20 mmHg; therefore, the expected HCO₃− = 24 − (2 × 2) = 20 mmol/L.

If this were a chronic respiratory alkalosis, for every 10 mmHg decrease in Paco₂, HCO₃− should decrease by 5 mmol/L; thus, the expected HCO₃− = 24 − (5 × 2) = 14 mmol/L.

Discussion

This case represents acute respiratory alkalosis most likely from a pulmonary embolus. The ventilation–perfusion mismatch causes hypoxia. The increased dead space (areas ventilated but not perfused) causes impaired CO₂ elimination, triggering the chemoreceptors in the carotid and aortic bodies and the brainstem to induce hyperventilation. The treatment of respiratory alkalosis should involve correcting the underlying disorder (Box 4); there is no place for the use of respiratory depressants.

Vignette 3

A 20‐year‐old woman with no known past medical history presents with a sensation of dyspnoea. Her respiratory rate is 28 breaths per minute; an ABG test was performed to assess oxygenation and ventilation: pH 6.93, Paco₂ 28 mmHg, HCO₃− 2 mmol/L, sodium (Na⁺) 147 mmol/L (RI, 135–145 mmol/L), potassium (K⁺) 4 mmol/L (RI, 3.3–4.5 mmol/L), chloride (Cl⁻) 111 mmol/L (RI, 96–106 mmol/L). There is a very severe acidaemia (low pH) with respiratory alkalosis (low Paco₂) and metabolic acidosis (low HCO₃−). Low HCO₃− with acidaemia indicates that metabolic acidosis is the predominant or primary process.

The expected compensatory Paco₂ is calculated by the formula Paco₂ = 1.5 (HCO₃− + 8) = 1.5 (2 + 8) = 15 mmHg. This patient's CO₂ level is higher than predicted, indicating an additional respiratory acidosis is present. This is likely due to respiratory muscle fatigue induced by the very high minute ventilation needed to compensate for the severe metabolic acidosis.

The next step (Box 5) is to calculate the anion gap (AG) (RI, 8–16 mmol/L): AG = Na⁺ + K⁺ − Cl⁻ −  HCO₃− = 147 + 4 − 111 − 2 = 38 mmol/L (high).

Then check for the presence of multiple metabolic disorders by calculating the delta gap (delta HCO₃− [DHCO₃− ] = 24 − measured HCO₃−, and delta AG = measured AG − 16). In this case, DHCO₃−  = 24 − 2 = 22 mmol/L, and delta AG = 38 − 16 = 22 mmol/L.

The two deltas are equal; therefore, there are no other metabolic processes besides high AG metabolic acidosis (HAGMA).

Discussion

This patient's tachypnoea was not due to a primary respiratory problem, but rather from respiratory compensation for the severe HAGMA. This patient's lactate and renal function were normal; however, her blood glucose was 44 mmol/L (RI, 3.5–5 mmol/L) with ketones of 7 mmol/L (RI < 0.6 mmol/L), so her HAGMA is due to diabetic ketoacidosis. If toxins are suspected, further history, examination, targeted investigations and discussion with specialist toxicologists will assist in making a diagnosis (Box 6).

An ABG analysis was needed to identify that this patient is unable to maintain the required minute ventilation for full respiratory compensation. Treatment should be aimed at rapid reversal of both the metabolic and respiratory acidosis.

Metabolic acidosis is defined by an excess of acid, or a deficit of base in the blood. The causes of metabolic acidosis (Box 6) can be divided into those with a high (> 16) versus a normal (≤ 16) AG: AG = (Na⁺ + K⁺) − (Cl⁻ + HCO₃−) (in mmol/L).

The AG should be corrected to account for a low albumin: corrected AG = AG + 0.25 (40 g/L − measured albumin).

The compensation for metabolic acidosis is a respiratory alkalosis, which occurs within minutes — as CO₂ is controlled by minute ventilation. Maximal respiratory compensation achieves a CO₂ level of about 10 mmHg; however, this degree of respiratory effort is not sustainable for prolonged periods of time.

Some patients may have several coexisting metabolic processes. It is important to identify all the underlying disorders to optimally treat the patient. This is done by comparing the DHCO₃− (DHCO₃− = 24 − measured HCO₃− = the total sum of all metabolic processes occurring) with the delta AG (delta AG = calculated AG − 16 = the HAGMA component):

• DHCO₃− = delta AG: the acidosis is solely due to HAGMA;
• DHCO₃−> delta AG: there is an additional normal AG metabolic acidosis;
• DHCO₃−< delta AG: there is an additional metabolic alkalosis.

Vignette 4

A patient presents with 3 days of vomiting after eating take‐away food. Laboratory tests reveal a high HCO₃−. ABG analysis shows pH 7.52, Paco₂ 52 mmHg, HCO₃− 42 mmol/L.

There is alkalaemia (high pH) with respiratory acidosis (high CO₂) and metabolic alkalosis (high HCO₃−). The primary process is the metabolic alkalosis as this correlates with alkalaemia. The expected Paco₂ level is Paco₂ = 40 + 0.6 (HCO₃− − 24) = 40 + 0.6 (42 − 24) = 51 mmHg.

This patient has a compensated metabolic alkalosis.

Discussion

Metabolic alkalosis is defined by the presence of high HCO₃−, with the possible differential diagnosis presented in Box 7. Compensation for metabolic alkalosis is respiratory acidosis, with maximal compensation generating a Paco₂ of around 60 mmHg. The kidneys are normally able to efficiently regulate HCO₃− production and secretion. Therefore, metabolic alkalosis always requires both an increase of bases and an impairment of renal HCO₃− excretion.

Vignette 5

A patient presents with an altered conscious state to the emergency department. No history is available, so an ABG test is performed: pH 7.34, Paco₂ 23 mmHg,

HCO₃− 15 mmol/L, Na⁺ 147 mmol/L, K⁺ 7 mmol/L, Cl⁻ 95 mmol/L. There is very mild acidaemia (low pH) with respiratory alkalosis (low Paco₂) and metabolic acidosis (low HCO₃−). The primary process is the metabolic acidosis, which correlates with acidaemia. Expected Paco₂ = 1.5 (HCO₃− + 8) = 1.5 (15 + 8) = 34 mmHg. Expected respiratory compensation should generate a Paco₂ level of about 34 mmHg. The measured Paco₂ is lower than this, so there is concurrent respiratory alkalosis.

There is a HAGMA: AG = Na⁺ + K⁺ − Cl⁻ − HCO₃− = 147 + 7 − 95 − 15 = 44. The delta AG (= 44 − 16 = 28) is much higher than the DHCO₃− (24 − 15 = 9) indicating that there is a superimposed metabolic alkalosis.

Discussion

In this case, there are three concomitant disorders: respiratory alkalosis, HAGMA and metabolic alkalosis. The metabolic alkalosis was only discovered on calculation of the delta gap, illustrating the benefit for completing a structured approach in every patient. History, examination and targeted investigations are needed to suggest a cause. In this patient, the diagnosis is severe pneumonia with high lactate, hyperventilation and diuretic therapy.