a. All acid-base disorders should be placed into one of the following categories:
i. Metabolic acidosis
1. Nonanion gap
2. Anion gap
ii. Metabolic alkalosis
1. Chloride responsive
2. Chloride unresponsive
iii. Respiratory acidosis
iv. Respiratory alkalosis
b. Often, there is more than one metabolic derangement, which can make interpretation difficult. Indeed, virtually any combination of the previous disorders can occur in an individual patient. However, respiratory acidosis and alkalosis cannot concurrently exist in the same patient. The approach discussed in this chapter will enable you to diagnose most acid-base disorders.
B. Steps in the Evaluation of Acid-Base Disorders.1 An arterial blood gas and electrolyte panel are needed to fully evaluate all acid-base disorders.
a. Decide whether the patient is acidemic or alkalemic. Although a pH of 7.35–7.45 is considered “normal,” mixed disorders or the body’s compensatory mechanisms may hide significant acid-base derangements within this range. If the clinical context suggests presence of an acid-base disorder, it is useful to decide on acidemia or alkalemia on the basis of whether the pH falls below or above 7.40.
i. Acidemia is diagnosed when the pH is less than 7.40.
ii. Alkalemia is diagnosed when the pH is greater than 7.40.
b. Determine whether the acid-base abnormality has a metabolic or respiratory cause. You can make this determination by looking at the arterial carbon dioxide tension (PaCO2). A high PaCO2 and a low pH indicate respiratory acidosis, whereas a high PaCO2 and a high pH indicate metabolic alkalosis. A low PaCO2 and a low pH indicate metabolic acidosis, and a low PaCO2 and a high pH indicate respiratory alkalosis.
1. If the patient is acidemic and the PaCO2 is high, the patient has respiratory acidosis. An acute respiratory acidosis will cause serum bicarbonate to rise about 1 mEq/L per 10 mm Hg rise of PaCO2. After 6–12 hours, the kidney can begin to generate more bicarbonate, and over days a patient can develop chronic respiratory acidosis with serum bicarbonate increasing about 3 mEq/L per 10 mm Hg rise in PaCO2.
2. If the patient is acidemic and the PaCO2 is low, you have diagnosed metabolic acidosis (because the PaCO2 does not account for the acidosis). Pure metabolic acidosis results in about a 1–1.3 mm Hg decrease in PaCO2 per 1 mEq/L decrease in serum bicarbonate.
1. If the patient is alkalemic and the PaCO2 is low, the patient has respiratory alkalosis. Acute respiratory alkalosis will cause serum bicarbonate to fall about 2 mEq/L per 10 mm Hg fall in PaCO2. Chronic respiratory alkalosis will cause serum bicarbonate to fall about 5 mEq/L per 10 mm Hg fall in PaCO2.
2. If the patient is alkalemic and the PaCO2 is high, you have diagnosed metabolic alkalosis. Pure metabolic alkalosis results in a 0.7 mm Hg increase in PaCO2 for each 1 mEq/L increase in serum bicarbonate.
c. Determine appropriate compensation and assess for secondary disorders. After the primary disorder has been identified, the expected compensation should be calculated. If the patient has not compensated as expected, there is likely a secondary (or secondary and tertiary) acid-base disturbance present.
For every change of 10 mm Hg in the PaCO2 (up or down), the pH changes 0.08 if the process is acute and 0.03 if the process is chronic (in the opposite direction of the PaCO2).
i. Respiratory acidosis: if increase in HCO3– is less than 1 mmol/L per 10 mm Hg rise in PaCO2, there is an additional metabolic acidosis. If increase in HCO3– is greater than 5 mmol/L per 10 mm Hg rise in PaCO2, there is an additional metabolic alkalosis.
ii. Metabolic acidosis: expected PaCO2 = 1.5 . HCO3– + 8 ± 2 (also called Winters’ formula). If higher, there is an additional respiratory acidosis. If lower, there is an additional respiratory alkalosis.
iii. Respiratory alkalosis: if decrease in HCO3– is greater than 5 mmol/L per 10 mm Hg rise in PaCO2, there is an additional metabolic acidosis. If decrease in HCO3– is less than 2 mmol/L per 10 mm Hg decrease in PaCO2, there is an additional metabolic alkalosis.
iv. Metabolic alkalosis: expected PaCO2 = 0.7 . (HCO3– – 24) + 40 ± 2. If higher, there is an additional respiratory acidosis. If lower, there is an addition respiratory alkalosis.
d. Calculate the anion gap. The anion gap equals the measured cations minus the measured anions; that is, Na+ – (Cl– + HCO3–). Because the measured cations are normally more than the measured anions, the unmeasured anions must be greater than the unmeasured cations by exactly the same amount to maintain electroneutrality. Any disorder that increases unmeasured cations (acids) will decrease measured anions and cause an increased anion gap.
i. Normal anion gap is 8–12.
ii. If the anion gap is more than 20, anion gap acidosis is present. The presence of an anion gap acidosis always represents a primary abnormality (see C. a. ii.).
iii. If the anion gap is 12–20, an underlying anion gap acidosis might still exist.
e. Calculate the corrected serum bicarbonate
i. The purpose of this calculation is to determine what the serum bicarbonate would be if no anion gap existed (i.e., the corrected serum bicarbonate).
1. If correcting the anion gap results in an elevated serum bicarbonate (i.e., >28 mmol/L), the patient has an underlying metabolic alkalosis.
3. If correcting the anion gap results in a normal serum bicarbonate, then the decreased serum bicarbonate is completely explained by the anion gap acidosis.
ii. The following formula can be used to calculate the corrected serum bicarbonate:
CB = Measured AG – Normal AG + Measured HCO3–
(CB = corrected bicarbonate and AG = anion gap)
By subtracting the normal anion gap from the measured anion gap, you have an estimate of the “extra acid” that is present. Because each extra acid titrates approximately one base, this calculation approximates the amount of bicarbonate consumed in titrating the anion gap acidosis. By adding this value to the measured bicarbonate value, you correct the bicarbonate for the effect of the anion gap acidosis.
Steps for Determining Acid-Base Abnormalities:
1. Determine whether the pH is greater than or less than 7.4
2. Look at the PaCO2
3. Determine the expected compensation
4. Calculate the anion gap
5. Calculate the corrected serum bicarbonate
C. Differential Diagnoses. After you have completed the five steps described in B., you will have identified most possible acid-base disorders. Three disturbances simultaneously (the “triple ripple”) is the maximum because respiratory alkalosis and respiratory acidosis cannot exist simultaneously. The following differentials can help you arrive at a cause for the patient’s acid-base disorder or disorders.
a. Metabolic acidosis
i. Nonanion gap. The common causes of nonanion gap acidosis include two renal, two gastrointestinal, and two “post” causes.
a. Acute or chronic kidney disease usually causes a mixed anion and nonanion gap acidosis.
b. Renal tubular acidosis (see Chapter 40)
a. Diarrhea. Bicarbonate loss can result in a nonanion gap metabolic acidosis.
b. Colovesicular fistula or ileostomy can also cause bicarbonate loss.
3. “Post” disorders
a. Post-hyperventilation. The kidney compensates for a respiratory alkalosis by “spilling” bicarbonate to generate a nonanion gap metabolic acidosis. If the respiratory alkalosis ceases (e.g., after treatment of heart failure), a nonanion gap acidosis may remain until the kidney can regenerate bicarbonate.
b. Post–anion gap acidosis. This can occur typically following correction of diabetic ketoacidosis due to preferential reabsorption of chloride in the proximal tubule and urinary loss of the unmeasured anion (which would normally be used to regenerate bicarbonate). Post–anion gap acidosis improves with time as the kidney regenerates bicarbonate.
Methanol intoxication (through conversion into formic acid)
Uremia (urea is an anion)
Diabetic or alcoholic ketoacidosis
Paraldehyde (a medicine no longer in use)
Isoniazid or Iron overdose
Lactate (usually from anaerobic metabolism during shock or extensive tissue injury)
Ethylene glycol intoxication (antifreeze ingestion)
1. Uremia. Check the creatinine and blood urea nitrogen (BUN).
2. Diabetic ketoacidosis. Diabetic ketoacidosis can be ruled out on the basis of a negative urine dipstick for ketones.
3. Salicylate overdose. A salicylate level should be obtained for an unexplained anion gap acidosis.
4. Methanol or ethylene glycol intoxication. A simultaneous blood sample for osmolarity, serum electrolytes, and ethanol level is especially important in patients with altered consciousness and in alcoholic patients with anion gap acidosis because the probability of a methanol or ethylene glycol ingestion is increased.
a. Calculate the osmolar gap. All alcohols (including ethanol, isopropyl alcohol, methanol, and ethylene glycol) can produce an osmolar gap (a difference between measured and calculated osmolarity), but only methanol and ethylene glycol lead to osmolar gaps with significant anion gap acidosis. Calculate the osmolar gap as follows:
Calculated osmolarity = 2 × Na + BUN/2.8 + glucose/18
Osmolar gap = Actual osmolarity – Calculated osmolarity
b. Correct for ethanol. The ethanol level divided by 4.6 equals the amount of osmoles that ethanol is contributing to the gap.
c. Calculate the remaining osmolar gap. Subtract the osmoles due to ethanol from the original osmolar gap. The remaining osmolar gap is still unexplained. A value greater than 5 mOsm/kg H2O is concerning (but not specific) for a toxic ingestion of another alcohol (e.g., methanol or ethylene glycol). Other clinical evidence that may confirm your suspicion includes visual disturbances (with methanol ingestion) and urinary oxalate crystals (with ethylene glycol ingestion).
5. Lactate. Often lactic acidosis is diagnosed on a clinical basis (e.g., obvious sepsis) or by exclusion. An elevated lactate level confirms your clinical suspicion.
6. Alcoholic ketoacidosis is common but should be a diagnosis of exclusion so that other potentially treatable and life-threatening conditions are not missed in patients with alcoholism.
b. Metabolic alkalosis. To determine the cause of metabolic alkalosis, first obtain a urine chloride level. Causes are referred to as chloride responsive or chloride unresponsive.
1. Prerenal states (e.g., severe heart failure). Most metabolic alkalosis is generated by the kidney reacting to a decrease in renal blood flow. Any of the prerenal states (see Chapter 37) can lead to metabolic alkalosis by:
a. Increased proximal bicarbonate reabsorption. Avid reabsorption of sodium in the proximal tubule induces increased bicarbonate reabsorption in order to maintain electroneutrality.
b. Increased distal acid secretion. A prerenal state results in higher renin and aldosterone levels, which increases sodium uptake in the distal tubule while leading to potassium and hydrogen ion secretion. Acid secretion equates to bicarbonate generation.
2. Gastric fluid loss (from vomiting or a nasogastric tube) can lead to a metabolic alkalosis.
3. Prior diuretic therapy leading to volume depletion can result in metabolic alkalosis.
ii. Urine chloride >20 mEq/L. Metabolic alkalosis resulting from the following causes will not correct with sodium chloride administration. The most common causes can be remembered by the mnemonic “ABCD”:
Depletion of magnesium
1. Primary hyperaldosteronism increases excretion of potassium and hydrogen ion at the distal tubule, resulting in hypokalemia and alkalosis.
2. Bartter’s syndrome results from defects in salt reabsorption in the thick ascending limb of the renal tubule. Salt and water loss trigger aldosterone production, which causes hypokalemia and alkalosis (hypomagnesemia may also be seen).
3. Cushing’s syndrome may produce hypokalemia and alkalosis by mechanisms similar to those of primary hyperaldosteronism.
4. Depletion of magnesium may result in potassium wasting. Hypokalemia may cause hydrogen ions to shift intracellularly and may also result in increased excretion of hydrogen at the distal tubule; both mechanisms may lead to alkalosis.
Bartter’s syndrome and magnesium depletion are generally associated with normal or low blood pressure, whereas primary hyperaldosteronism and Cushing’s syndrome are generally associated with hypertension.
Diuretics often produce a confusing picture because although urinary chloride is >20 mEq/L, metabolic alkalosis is usually responsive to sodium chloride administration.
a. Respiratory acidosis (hypoventilation). See Chapter 25.
b. Respiratory alkalosis (hyperventilation). There are seven common etiologies:
i. Primary central nervous system (CNS) disorders
ii. Pulmonary disease (including all causes of hypoxemia)
Respiratory alkalosis may be the first acid-base abnormality seen in patients with sepsis.
v. Drugs (e.g., salicylates)
vi. Liver disease
vii. Pain or anxiety
Suggested Further Readings
Berend K, de Vries APJ, Gans ROB. Physiological approach to assessment of acid-base disturbances. N Engl J Med 2014;371:1434–45.Find this resource:
Haber RJ. A practical approach to acid-base disorders. West J Med 1991;155:146–51. (Classic Article.)Find this resource:
1 Modified with permission from Haber RJ. A practical approach to acid-base disorders. West J Med 1991;155(2):146–151.