Bicarbonate deficit

Clinical background

In severe metabolic acidosis the question arises whether intravenous bicarbonate should be given and, if so, how much. The decision is difficult for two reasons. First, there is no clear evidence that bicarbonate replacement improves survival in the commonest forms of acute metabolic acidosis, above all lactic acidosis and diabetic ketoacidosis [4, 5]. Second, bicarbonate has no fixed volume of distribution in the body, unlike electrolytes that do not participate in buffer systems. The apparent volume of distribution, often called the bicarbonate space, varies with the degree of acidaemia and makes a simple dose calculation uncertain [1, 2]. The calculator addresses the second problem: it gives an estimate of the amount of bicarbonate required to reach a chosen target value, based on a volume of distribution adjusted for the severity of the acidaemia.

Calculating the bicarbonate deficit

The formula rests on the principle that the amount of bicarbonate needed is the product of body weight, volume of distribution and the difference between the desired and the measured bicarbonate:

$$\text{HCO}_3^-\text{ deficit (mmol)} = 0{.}5 \times \text{weight (kg)} \times (\text{desired HCO}_3^- - \text{measured HCO}_3^-)$$

The factor 0.5 represents the apparent volume of distribution in L/kg and derives from Adrogué and Madias's review of the management of life-threatening acid–base disorders, published in the New England Journal of Medicine in 1998 [1]. There the authors describe how the bicarbonate space is not a fixed physical volume but expands as the acidaemia worsens, because hydrogen ions buffered in non-bicarbonate buffer systems are released and consume administered bicarbonate. In moderate metabolic acidosis (HCO3 approximately 17 to 20 mmol/L) the volume of distribution is estimated at approximately 0.5 L/kg, the factor the calculator uses as default. In severe acidaemia (HCO3 below approximately 10 mmol/L) the apparent volume rises towards 0.7 to 0.8 L/kg, which means that the formula underestimates the requirement in this range if the factor 0.5 is retained unchanged [1, 2].

Repetto and Penna studied the apparent bicarbonate space in children with metabolic acidosis and found that it varied markedly with the starting HCO3 [2]. At low starting values the space was considerably greater than 0.5 L/kg, confirming that a fixed factor systematically underestimates the dose requirement in severe acidaemia. Their data support a stepwise model in which the volume of distribution increases as the HCO3 falls, but clinical application remains crude because inter-individual variation is large.

Interpretation in practice

The calculator gives an estimate of the total amount of bicarbonate in mmol that would in theory raise the serum HCO3 to the desired value. In practice the dose should never be given as a single bolus up to the full calculated amount. The table below summarises how the result should be handled.

The target should always be set at partial correction, not normalisation. Aiming for an HCO3 of approximately 15 mmol/L or a pH > 7.20 is an established strategy that minimises the risks of overcorrection while limiting the acute haemodynamic consequences of severe acidaemia [1, 3]. Full normalisation of the HCO3 is not a treatment goal in acute metabolic acidosis.

Validation and performance

The formula is not a validated risk stratification instrument in the traditional sense, and there is no external validation cohort with a reported c-statistic or calibration. Its ability to predict the actual dose required to reach a given HCO3 target has been tested indirectly in clinical studies of the size of the bicarbonate space.

Repetto and Penna measured the apparent bicarbonate space in children given a bicarbonate infusion for metabolic acidosis and compared it with adult data from the literature [2]. They found that the space varied from approximately 0.5 L/kg in moderate acidaemia to considerably higher values in severe acidaemia, with wide individual scatter. This means that the formula with the factor 0.5 may underestimate the requirement by up to 40 per cent in severe acidaemia (HCO3 below 10 mmol/L), where the actual volume of distribution approaches 0.7 to 0.8 L/kg.

The only large randomised study of bicarbonate in severe metabolic acidaemia is BICAR-ICU, a multicentre study from 26 French intensive care units with 389 adult patients (pH ≤ 7.20, HCO3 ≤ 20 mmol/L) [5]. The study found no effect on the composite primary outcome (death by day 28 or organ failure at day 7) in the population as a whole. In a prespecified subgroup of patients with acute kidney injury (AKIN 2 to 3), however, 28-day survival was significantly higher with bicarbonate (54 per cent versus 37 per cent, p = 0.028). The study did not use the formula for dosing but gave 4.2% bicarbonate solution 125 to 250 mL per infusion to keep the pH above 7.30, up to 1000 mL per day, which is a more conservative strategy than an uncorrected calculation would give.

A French expert panel (SRLF and SFMU) published guidelines in 2019 on the diagnosis and management of metabolic acidosis using GRADE methodology [3]. The panel defines severe acidaemia as a pH ≤ 7.20 and recommends that bicarbonate may be considered in severe metabolic acidaemia, particularly with concurrent acute kidney injury, but stresses that treatment of the underlying cause is always primary.

Limitations

The formula does not apply to all forms of metabolic acidosis. In diabetic ketoacidosis most guidelines do not recommend routine bicarbonate replacement even at a pH below 7.20, since insulin treatment rapidly switches off ketone production and bicarbonate may delay ketone clearance [4]. In lactic acidosis from sepsis or shock the evidence is weak and bicarbonate should be considered only when the pH is very low and ventilation is not compensating [4, 5].

The most important pitfalls in using the formula are:

Overestimating its precision. The formula gives a rough estimate, not an exact dose. The apparent volume of distribution varies with the severity of the acidaemia and with the individual's buffering capacity. In severe acidaemia (HCO3 below 10 mmol/L) the formula underestimates the requirement, and in moderate acidaemia it may overestimate it. Dosing must always be verified with repeated blood gases.

Overcorrection. Giving the whole calculated dose at once risks metabolic alkalosis, hypernatraemia and ionised hypocalcaemia. The BICAR-ICU study reported metabolic alkalosis, hypernatraemia and hypocalcaemia significantly more often in the bicarbonate group [5]. Hypocalcaemia can prolong the QTc interval and increase the risk of arrhythmia [4].

Paradoxical intracellular and CNS acidosis. Bicarbonate is metabolised to carbon dioxide, which diffuses across the blood–brain barrier faster than bicarbonate. This can lower the pH of the cerebrospinal fluid despite a rising arterial pH, with neurological consequences [4]. The risk is greater if the patient cannot increase minute ventilation sufficiently to blow off the additional carbon dioxide, for example during sedation or neuromuscular blockade.

Sodium load. Bicarbonate is given as a sodium salt. With impaired renal function or heart failure, the sodium load accompanying a full dose according to the formula can cause volume overload. In these cases renal replacement therapy should be considered as an alternative or an adjunct.

The formula must not be used in respiratory acidosis, where the problem is carbon dioxide retention and not a bicarbonate deficit. Giving bicarbonate in respiratory acidosis worsens the hypercapnia, since bicarbonate is metabolised to carbon dioxide.

References

1. Adrogué HJ, Madias NE. Management of life-threatening acid-base disorders. First of two parts. N Engl J Med 1998. PMID: 9414329
2. Repetto HA, Penna R. Apparent bicarbonate space in children. TheScientificWorldJournal 2006. PMID: 16493519
3. Jung B, Martinez M, Claessens YE et al. Diagnosis and management of metabolic acidosis: guidelines from a French expert panel. Ann Intensive Care 2019. PMID: 31418093
4. Eraky AM, Yerramalla Y, Khan A et al. Complexities, Benefits, Risks, and Clinical Implications of Sodium Bicarbonate Administration in Critically Ill Patients: A State-of-the-Art Review. J Clin Med 2024. PMID: 39768744
5. Jaber S, Paugam C, Futier E 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. PMID: 29910040