Hypokalaemia and Metabolic Acidosis

35 year old Aboriginal female presents with a 2/52 Hx of weakness, thirst and nausea. Presents to ED unable to lift her hands. Admitted 3/12 ago with something similar but doesn’t know what it was and her medical notes are not immediately available. No other past medical history of note. Examination reveals a quiet, dehydrated lady with generalised non-lateralising weakness in all 4 limbs.

HR 75 /min
BP 90/50 mmHg supine
Temp 37.1 oCelsius

Bedside venous blood gas results included:

Reference Range
FIO2 0.21
pH 7.25 7.35-7.45
pCO2 35 mmHg 35-45
pO2 234 mmHg 80-95
HCO3 15.0 mmol/L 22-28
Base Excess -11 -3 to +3
Na+ 141 mmol/L 134-146
K+ 1.1 mmol/L 3.4-5.0
Cl 116 mmol/L 98-106
Glucose 5.4 mmol/L 3.5-4.5


This is the ECG on presentation:

Hypokalaemia and metabolic acidosis



Metabolic acidosis: low pH + normal CO2 + normal HCO3 + strongly negative base excess.

Na – [HCO3 – Cl] = 141 – 15 – 116 = 10

  • Sinus rhythm with sinus arrhythmia at a rate of 72 bpm.
  • Normal axis.
  • Unremarkable PQRST complexes.
  • U waves noted – most prominently in leads V1-V3
  • Sinus arrhythmia [sinus rhythm with slight variation (>0.16 seconds) in the sinus cycles]

  • Normal anion gap metabolic acidosis. The 2 most common causes in ED
    • Diarrhoea (>95%)
    • Renal tubular acidosis
  • Other causes are many and varied. There are several mnemonics out there – the most recent edition of Rosen suggests: F-USED CARS
F Fistula (Pancreatic)
U Uretoenteric conduits
S Saline administration XS
E Endocrine (Hyperparathyroid)
D Diarrhoea
C Carbonic anhydrase inhibitors
A Arginine, lysine, chloride (TPN excess)
R Renal tubular acidosis
S Spironolactone


  • Basically (and rather obviously), a metabolic acidosis is caused by either excess acid or a loss of alkali.
  • Excess acid may be produced by the body itself or may be exogenous.
  • Calculating the anion gap is used in the context of having made a diagnosis of a metabolic acidosis to help determine possible causes.
  • It’s an artificial but pragmatic concept based on the fact that with normal physiology there will be more unmeasured anion’s (predominantly Albumin, Phosphate and Sulphate) than cations on routine blood testing.

The most commonly used formula is:

AG = Na – HCO3 – Cl

Most people don’t use potassium in the equation – resulting in a normal range of 8-12. (12-16 if potassium included), although in this case it wouldn’t have made much difference!

A wide anion gap in the setting of a metabolic acidosis (or High Anion Gap Metabolic Acidosis [HAGMA]) suggests there is excess unmeasured anion / acid. Keeping it simple, there are only 4 causes:

  • Lactic acidosis
  • Ketoacidosis
  • Uraemia
  • Exogenous Acid

A commonly used mnemonic for a wide anion gap acidosis is CAT MUDPILES

C Carbon Monoxide, Cyanide
A Alcohol, Alcoholic Ketoacidosis
T Toluene
M Methanol, Metformin
U Uraemia
D Diabetic Ketoacidosis (acute)
P Paraldehyde, Propylene Glycol
I Iron, Isoniazid
L Lactate
E Ethylene Glycol
S Salicylates

Essentially a state of excess HCO3 loss or loss of ability to excrete H+. An elevated chloride is a hallmark feature – in fact a normal anion gap metabolic acidosis is sometimes referred to as a hyperchloraemic acidosis. More than 95% of cases of a normal anion gap metabolic acidosis in ED’s are caused by diarrhoea. The second most common cause is Renal Tubular acidosis.

This patient, it turns out had been diagnosed with renal tubular acidosis 3 months previously on the basis of presenting with a normal anion gap metabolic acidosis in the setting of hypokalaemia with no overt GI loss.

Renal Tubular Acidosis

  • Essentially a failure of H+ excretion or HCO3 reabsorption in the renal tubules.
  • There are 4 (3) types, but the classification has little utility in Emergency medicine and therefore won’t be discussed.
  • Causes of renal tubular acidosis are legion but common ED causes include:
    • Chronic renal disease
    • Genetic
    • Autoimmune disease
    • Drugs: such as Lithium, Ibuprofen
    • Toluene abuse
  • Management of renal tubular acidosis involves supportive therapy for any acute presentation followed by long term oral bicarbonate therapy.
  • This patient had been taking 4g of ibuprofen a day for a protracted period for toothache and this had been mooted to be the cause. There was no obvious cause for her recurrence.

This patient had the lowest potassium I have ever seen…

  • 98% of the body’s potassium is intracellular and it is the major ion contributing to osmotic and electrochemical gradients in the body.
  • The balance between intra and extracellular potassium is largely maintained by the NA K ATPase pump.
  • Insulin and Beta-adrenergic agonists promote intracellular movement.
  • Aldosterone shifts potassium into the extracellular space, but then promotes its renal excretion.
  • Acidosis promotes potassium movement out of the cell in exchange for an H+ ion therefore leading to hyperkalaemia.


  • Renal Losses:
    • Diuretics
    • Hyperaldosteronism eg: CCF, liver failure
    • Renal tubular acidosis
  • Extra-renal Depletion:
    • Malnutrition – eg: alcoholism, eating disorders, ‘tea + toast’ diet in elderly
    • Vomiting
  • Re-distributional:
    • Alkalosis
    • Insulin in DKA therapy
    • Beta-agonists.
  • Pseudo-hypokalaemia:
    • If sample exposed to a high ambient temperature before analysis.

  • As hyperkalaemia tends to cause bradycardia and eventually asystole, HYPOkalaemia tends to make the myocardium more irritable and prone to Ventricular Fibrillation (VF).
  • As VF is felt to be the primary arrhythmia in >90% of sudden deaths, many cardiologists prefer their patients to run a high-normal serum potassium. Indeed, in the setting of a low-normal potassium and an arrhythmia that is proving difficult to control, many advocate a potassium infusion as part of the arrhythmia management.

Classical ECG changes in Hypokalaemia include:

  • T wave flattening or inversion
    • Imagine the T waves are actually made of potassium – they increase in size as potassium rises!
  • U waves
    • Often mistaken for P waves.
    • Called U waves because they follow the T wave, not because they’re shaped like a ‘U’!
    • Often fused with the T wave and leading to the illusion of QT prolongation.
    • Purists state this is NOT a cause of QT prolongation but pragmatists point out that they are prone to the same arrhythmias as people with other true causes of a prolonged QT interval. I’m a pragmatist.
    • Partial fusion leads to a distinctive camel hump (seen in V3 in this patient’s ECG).
    • U waves are most commonly seen in the setting of hypokalaemia but they can be a normal variant and are also seen in ischaemia, hypercalcaemia, sodium channel blockade, hyperthyroidism + digoxin toxicity, to name a few!
  • Atrial Fibrillation
  • Ventricular tachycardia
  • Torsades de Pointes (TDP)
    • TDP is a variant of Polymorphic Ventricular tachycardia and technically requires there to be underlying marked QT prolongation (usually >600msec)

Hypokalaemia and metabolic acidosis

  • Oral replacement is reasonable with a K >2.5 mmol/l unless there are severe associated symptoms (such as arrhythmias or respiratory muscles weakness).
  • Below 2.6 mmol/l, IV is the preferred method of replacement. With so much of the body’s potassium being intracellular its hard to gauge how much is required but a rough guide to the estimated amount of potassium the patient will need to have replaced is that they will need 100 mmol for every 0.3 mmol/l below the normal serum potassium cut-off – in this case the patients potassium was about 2.3 mmol/L below the lower cut off and therefore would roughly require nearly 800 mmol!
  • One of the key things that people forget when dealing with hypokalaemia is to co-administer magnesium. Hypokalaemia is almost invariably linked with hypomagnesaemia. Magnesium exerts an important effect on the Na K ATPase pump and having a low magnesium promotes the loss of potassium out of the cell. Therefore giving potassium in isolation is a very ineffecient method of replacement – the majority will just be pissed out.

How much magnesium should you give?

  • 10 mmol in the first hour and 5 mmol/hr afterwards is reasonable (with appropriate monitoring)

How fast can you give potassium?

  • Peripheral veins can usually tolerate up to 20 mmol/hr, thought this ay be extremely painful. In emergent cases can be given at a rate of up to 40mmol/hr. In cardiac arrest it’s reasonable to give 20mmol over 2-3 minutes and to repeat until the K is >4.0.
  • In this patient we decided early on that she would need so much potassium that a central line was the most reasonable course of action. She had her potassium corrected over 24 hours, recommenced her HCO3 and was discharged after 5 days having made a full recovery.

Print Friendly


  1. nickecia campbell says

    This case based presentation is eye opening: shows that every and anything comes to the ED and we need to put it all together. Thank u. Nickecia

  2. Ben says

    Hi Mike and co! The PO2 seems very high at 234mmHg with an FiO2 of 0.21. Is this right? Was there an oxygen bubble in the ABG? Hyperbaric Oxygen? Cheers