Bicarbonate Deficit Calculator
Bicarbonate Deficit Calculator
Estimate the bicarbonate needed to reach a target level, and see why bicarbonate replacement remains a genuinely controversial therapy.
Bicarbonate Deficit
3 inputs → deficit70 kg, actual bicarbonate 10 mmol/L, target 18 mmol/L
Formula
- 0.5
- the apparent bicarbonate distribution volume, as a fraction of body weight — an approximation that rises in severe acidosis, reaching 0.8 or more below a bicarbonate of about 5
- target
- conventionally 18 mmol/L, not a normal value — over-correction carries its own risks
Worked example
70 kg, actual bicarbonate 10 mmol/L, target 18 mmol/L
18 − 10 = 8 mmol/L to correct
0.5 × 70 × 8 = 280 mmol HCO₃⁻
Give no more than half of this initially and re-measure
How the distribution volume changes
| Bicarbonate level | Apparent distribution volume |
|---|---|
| Near normal | ≈ 0.4 × body weight |
| Moderate acidosis | ≈ 0.5 × body weight (the value used here) |
| Bicarbonate below ≈ 5 mmol/L | 0.8 × body weight or more |
A controversial correction, cautiously applied
Bicarbonate does not distribute solely through extracellular fluid; as pH falls, buffering draws in intracellular and bone stores as well, so the apparent volume of distribution rises with the severity of the acidosis — from roughly 0.4 of body weight when bicarbonate is near normal to 0.8 or more once it falls below about 5 mmol/L. The 0.5 constant used here is a fixed, moderate-acidosis approximation, and the formula understates the true requirement in the sickest patients precisely when accuracy matters most. In practice, give no more than half the calculated deficit as a first dose, then re-measure — both to reassess the true need and because the underlying cause of the acidosis may be resolving on its own.
The target of 18 mmol/L, not a normal value of around 24, is deliberate: over-correction produces hypokalaemia as potassium shifts intracellularly, hypocalcaemia from a rise in ionised calcium binding, and a paradoxical worsening of CSF acidosis, since CO₂ generated by the buffering reaction crosses the blood-brain barrier faster than bicarbonate itself.
Whether to give bicarbonate at all is genuinely contested and this calculator estimates a dose, not a recommendation to treat. It is not advised in diabetic ketoacidosis above a pH of 6.9. In lactic acidosis, evidence of benefit from the BICAR-ICU trial was confined to patients with severe acidaemia and concurrent acute kidney injury; outside that group, routine bicarbonate replacement has not been shown to improve outcomes and treating the underlying cause of the acidosis remains the priority.
Frequently asked questions
How is the bicarbonate deficit calculated?
Deficit = 0.5 × body weight (kg) × (target bicarbonate − actual bicarbonate), using an assumed distribution volume of half body weight.
Why does the formula use 0.5 rather than total body water?
0.5 approximates bicarbonate’s apparent distribution volume at a moderate acidosis. That volume actually rises as acidosis worsens, reaching 0.8 or more below a bicarbonate of about 5, so the formula underestimates the true deficit in severe cases.
Why is the target 18 and not a normal bicarbonate?
To avoid over-correction. Restoring bicarbonate too far or too fast risks hypokalaemia, hypocalcaemia and a paradoxical worsening of CSF acidosis as CO₂ crosses the blood-brain barrier faster than bicarbonate.
Is bicarbonate replacement recommended in DKA or lactic acidosis?
Not routinely. It is not recommended in diabetic ketoacidosis above pH 6.9. In lactic acidosis, the BICAR-ICU trial found benefit only in patients with severe acidaemia and acute kidney injury; treating the underlying cause remains the priority in other cases.
Related calculators
References
- Kraut JA, Madias NE. Treatment of acute metabolic acidosis: a pathophysiologic approach. Nat Rev Nephrol. 2012;8(10):589–601.
- Jaber S et al. Sodium bicarbonate therapy for patients with severe metabolic acidaemia in the intensive care unit (BICAR-ICU): a multicentre, open-label, randomised controlled, phase 3 trial. Lancet. 2018;392(10141):31–40.