Whenever the plasma bicarbonate rises above 24mmols/l, bicarbonate is excreted by the kidney. This response is reasonably prompt and effective so a metabolic alkalosis will be rapidly corrected. If you infuse say 100mls of 8.4% sodium bicarbonate into a healthy person with normal renal function, the rise in plasma bicarbonate is brief because of prompt bicarbonaturia. This is one way to alkalinise the urine. An infusion of alkali causes only a brief metabolic alkalosis due to this rapid renal excretion.
This ability of the kidney to rapidly excrete bicarbonate if its level is high is in complete contrast to its powerful ability to reabsorb all of the filtered load if plasma [HCO3] is low or normal. A useful analogy here is to filling a bucket. No water is lost until the bucket is full, but after that, all extra water is lost. This is sometimes called a waterfall effect.
The persistence of a metabolic alkalosis requires an additional process which acts to impair renal bicarbonate excretion. In our analogy, this would be something that increased the height of walls of the bucket. This means that two issues must to be considered when analysing a metabolic alkalosis:
When discussing the 'cause' of a metabolic alkalosis, note this term is used in several ways. For example it may be used to describe the initiating process, or the process maintaining the disorder or it can be used to refer to the combination of both processes, so be mindful of this when reading the rest of this section as otherwise you may become a little confused.
Normally, plasma bicarbonate is kept at a steady level of about 24 mmols/l by two renal processes:
Causes of a metabolic alkalosis can be classified into several groups as outlined in the table.
'Causes' : Classification of Initiating Processes for Metabolic Alkalosis
Gain of alkali in the ECF
Loss of H+ from ECF
Excessive intravenous administration of alkali alone will cause a metabolic alkalosis which is only short-lived because of rapid renal excretion of bicarbonate (as mentioned previously).
Hepatic metabolism of citrate, lactate, acetate or certain other organic acid anions to bicarbonate can cause a brief metabolic alkalosis. This may occur after a massive blood transfusion because of the metabolism of the administered citrate. The kidneys excrete the bicarbonate and the urine will be relatively alkaline.
This is discussed in section 7.3.
'Causes' of clinically significant chronic metabolic alkalosis are usefully divided into 2 major groupings based on the major factor involved in the maintenance of the disorder:
Maintenance of the alkalosis requires a process which greatly impairs the kidney's ability to excrete bicarbonate and prevent the return of the elevated plasma level to normal. Chloride deficiency leads to a situation where the kidney reabsorbs more bicarbonate anion than usual because there is not sufficient chloride anion present. Reabsorption of an anion is necessary to maintain electroneutrality as Na+ & K+ are reabsorbed so the deficiency of chloride leads to a re-setting upwards of the maintained plasma bicarbonate level. Chloride and bicarbonate are the only anions present in appreciable quantities in extracellular fluid so a deficiency of one must lead to an increase in the other because of the strict requirement for macroscopic electroneutrality.
Administration of chloride is necessary to correct these disorders. The four major sub-groups of metabolic alkalosis are listed in the table below. The two commonest causes of chronic metabolic alkalosis are loss of gastric juice and diuretic therapy. The gastric secretion of H+ results in generation of new bicarbonate which is returned to the blood.
Gastric alkalosis is most marked with vomiting due to pyloric stenosis or obstruction because the vomitus is acidic gastric juice only. Vomiting in other conditions may involve a mixture of acid gastric loss and alkaline duodenal contents and the acid-base situation that results is more variable. Histamine H2-blockers also decrease gastric H+ losses despite continued vomiting or nasogastric drainage and alkalosis will not occur if the fluid lost is not particularly acidic - indeed loss of alkaline small intestinal contents can even result in an acidosis if gastric acid secretion is suppressed.
Diuretics such as frusemide and thiazides interfere with reabsorption of chloride and sodium in the renal tubules. Urinary losses of chloride exceed those of bicarbonate. The patients on diuretics who develop an alkalosis are those who are also volume depleted (increasing aldosterone levels) and have a low dietary chloride intake ('salt restricted' diet). Hypokalaemia is common in these patients. If dietary chloride intake is adequate then an alkaosis is unlikely to develop. This is the main reason why every patient taking diuretics such as thiazides or lasix does not develop a metabolic alkalosis. The effect of diuretic use on urinary chloride levels depends on the relationship of the time of urine collection to diuretic effect: it is high while the diuretic is acting, but drops to low levels afterwards.
Villous adenomas typically excrete bicarbonate and can cause a hyperchloraemic metabolic acidosis. Sometimes they excrete chloride predominantly and the result is then a metabolic alkalosis.
Chloride diarrhoea is a rare congenital condition due to an intestinal transport defect, where the chronic faecal chloride loss can (if associated with volume depletion and K+ loss as maintenance factors) result in a metabolic alkalosis.
Potassium depletion occurs with situations of mineralocorticoid excess. Bicarbonate reabsorption in both the proximal and distal tubules is increased in the presence of potassium depletion. Potassium depletion decreases aldosterone release by the adrenal cortex.
A Common Hybrid Classification of 'Causes' of Metabolic Alkalosis
A: Addition of Base to ECF
B: Chloride Depletion
C: Potassium Depletion
D: Other Disorders
This condition is one cause of 'saline-resistant' metabolic alkalosis. The increased aldosterone levels lead to increased distal tubular Na+ reabsorption and increased K+ & H+ losses. The increased H+ loss is matched by increased amounts of renal HCO3- leaving in the renal vein. The net result is metabolic alkalosis with hypochloraemia and hypokalaemia, often with an expanded ECF volume.
The excess corticosteroids have some mineralocorticoid effects and because of this can produce a metabolic alkalosis. The alkalosis is most severe with the syndrome of ectopic ACTH production.
Cases have been reported of patients with metabolic alkalosis and severe hypokalaemia ([K+] < 2 mmol/l) due to severe total body potassium depletion. Investigation has not shown increased mineralocorticoid activity. The aetiology in these patients is not understood but correction of the alkalosis requires correction of the potassium deficit. These patients do not respond to saline loading unless K+ replacement is sufficient to correct the deficit. Urinary chloride losses are high (>20mmol/l).
This is a syndrome of increased renin and aldosterone levels due to hyperplasia of the juxtaglomerular apparatus 1,6. It is inherited as an autosomal recessive disorder. The increased aldosterone levels usually result in a metabolic alkalosis. The condition is usually found in children. Patients who present with hypokalaemic alkalosis of uncertain cause are often suspected of having this condition but other causes which may be denied by the patient should be considered eg surreptitious vomiting and/or use of diuretics for weight loss or psychological problems. These situations have been termed 'pseudo-Bartter's syndrome'. Rare genetic disorders such as Gitelmann's syndrome should also be considered.
Glycyrrhizin is precent in licorice root. It has a sweet taste with a licorine tang and is used in some countries (eg particularly Japan) as a food additive or in traditional medicines. It inhibits the conversion of cortisol to cortisone by inhibiting the enzyme 11-betahydroxysteroid dehydrogenase. The resulting high cortisol levels have a mineralocorticoid effect (pseudohyperaldosteronism) causing Na+ retention and excessive urinary K+ loss. Excessive intake may result in hypertension, oedema, hypokalaemia and metabolic alkalosis. 7
In most cases the cause is obvious (eg vomiting, diuretic use) but if not then measurement of a spot urinary chloride can be useful.
Two things to be aware of when interpreting the result:
(provided that recent thiazide use has been excluded).
If the clinical information is not sufficient to make a diagnosis the term 'idiopathic metabolic alkalosis' is sometimes used. The urinary chloride/creatinine ratio may occasionally be useful as it is elevated if there is an extra-renal cause of alkalosis.
Metabolic Alkalosis Based on Urinary Chloride
Urine Cl- < 10 mmol/l
Urine Cl- > 20 mmol/l