Blank SP, Blank RM, Laupland KB, Tabah A, Gill D, Kumar A, et al. Sodium bicarbonate administration for metabolic acidosis in the intensive care unit: a target trial emulation. Intensive Care Med. 2025 Jun;51(6):1–9.

 

The background

Critically ill patients often develop severe metabolic acidosis, associated with high mortality (1). Severe acidosis can potentially impact myocardial function, impair sensitivity to vasopressor support, and trigger malignant arrhythmias (2). The cornerstone of treatment is to address the underlying driver of acid generation. However, the administration of sodium bicarbonate might mitigate life-threatening adverse effects resulting from unrelenting metabolic acidosis, while allowing time for specific, targeted therapy to take effect (3). 

Despite the physiological rationale, results from clinical studies do not support the routine use of bicarbonate infusion – two small trials failed to reveal improvement in hemodynamic parameters (4,5). Besides, potential harmful effects can arise – including alkalosis, hypokalemia, hypernatremia, and intracellular acidosis due to diffusion of carbon dioxide into the cell (2). 

The BICAR-ICU randomized controlled trial did not reveal improvement in the composite primary outcome of mortality and organ dysfunction with bicarbonate administration. However, among patients with acute kidney injury (AKI), mortality was lower with bicarbonate therapy (5). 

Predictably, observational studies among critically ill patients have revealed conflicting results – as the sickest subgroup of patients usually receive bicarbonate therapy. Thorough analysis requires meticulous control of bias and appropriate adjustment for confounders – easier said than done. 

 

The methodology 

The Queensland Critical Care Research Network (QCCRN) study asked a big question – does sodium bicarbonate administration save lives among critically ill patients with metabolic acidosis (6)?

The authors hypothesized that the administration of sodium bicarbonate would lead to reduced ICU mortality among patients with metabolic acidosis. Performance of a randomized controlled trial to address this question might be a challenging undertaking; hence, as a first step, they performed a retrospective cohort study. 

They analyzed a large multicentric dataset to perform what they describe as a “target trial emulation” – the use of real-world data to mimic what a randomized controlled trial would look like, and simulate treatment effects. The retrospective dataset came from 12 ICUs in Queensland, Australia, collected between January 1, 2015 and December 31, 2021.

Observational data often carry major confounders. Sicker patients are more likely to receive bicarbonate, which can lead to bias when outcomes are analyzed. Hence, the authors used a method called G-computation to alleviate bias. In simple terms, it is a sophisticated statistical method of adjusting for differences between those who received bicarbonate and those who did not—not just at the start, but day by day, as their condition evolved over time. This method models how the risk of death changes over time, depending on whether they received bicarbonate or not. It accounts for daily changes in illness severity, including pH, lactate levels, and organ failure scores. 

For each day in the ICU, they collated data including pH, lactate level, SOFA score, and vasopressor use to model patient outcomes with and without bicarbonate. As the next step, they performed a simulation technique known as Monte Carlo simulation. This is basically a computer program that creates thousands of “what if” scenarios for each patient. In simple terms, it creates scenarios for what if a patient received bicarb? What if they didn’t? Who among them would survive and who would die?

They double-checked their findings with another statistical technique – “rolling entry matching”  – a nuanced form of propensity score matching (propensity score matching is a method that pairs patients with similar severity of illness but different treatments). Rolling entry matching was meant to augment the robustness of the results – to ensure that they were not due to chance or bad modeling.

In the end, this approach provides a much clearer picture of the potential benefit of bicarbonate—almost like a simulated randomized controlled trial. 

Study population 

Adult patients admitted to the ICU with metabolic acidosis – pH less than 7.3 with a PCO₂ less than 45 mm Hg were eligible, within the first 3 days of admission. 

Excluded 

• Dialysis-dependent renal failure
• Diabetic ketoacidosis
• Toxin ingestion
• Interhospital transfers  

 

Intervention

Administration of intravenous sodium bicarbonate on days when the eligibility criteria were present. The control group received no bicarbonate. 

Results 

Of 6157 eligible patients, 1764 (29%) received bicarbonate. Patients who underwent bicarbonate therapy were more severely ill (as expected), with higher APACHE III scores, more severe acidosis, a higher requirement for vasopressors, and renal replacement therapy (RRT). The unadjusted ICU mortality was significantly higher among those who received bicarbonate. 

Adjusted ICU mortality 

Using G-computation and Monte Carlo simulation, the study compared 3 treatment strategies. The findings were as follows: 

 

This represented a significantly lower absolute risk reduction (ARR) in ICU mortality with bicarbonate treatment of 1.9% (16.5% minus 14.4); 95% CI: 1.4–2.4%. Fifty-three patients would need to be treated with bicarbonate to save one life (the number needed to treat, NNT). 

NNT = 1 / ARR. Hence, for this study, the NNT is 1 / 0.019 = 52.63.

Sensitivity analysis 

The findings of G-computing were corroborated on rolling entry matching of pairs of patients. Overall, there were 1685 matched pairs of patients with similar severity of illness. The ICU mortality revealed an ARR of 2.1% (95% CI − 4.9 to 0.6) with bicarbonate therapy. The 30-day mortality also favored bicarbonate-treatment (ARR 3%, 95% CI − 5.9% to − 0.1%) (statistically significant). 

Subgroup analysis 

The favorable effect on ICU mortality was observed in subgroups of patients with severe acidosis (pH <7.2), acidosis with AKI, and among those on vasopressor support. No benefit was observed among patients with milder acidosis (pH >7.2), with no AKI or vasopressor support (See table)

Table: Subgroup analysis

 

 

Abbreviation: VIS: Vasoactive-inotropic score (a score of ≥ 10 refers indicates > 0.1mcg/kg/min norepinephrine)

Study conclusion 

Bicarbonate therapy was associated with a modest, but statistically significant (1.9%) reduction in ICU mortality. The findings suggest a causal benefit of early bicarbonate use in appropriately selected ICU patients with metabolic acidosis. The absolute benefit was greater among subgroups of patients with AKI, severe acidosis, and those requiring vasopressors. The mortality benefit was sustained to 30 days on sensitivity analysis. 

Strengths 

The authors employed a relatively new design, using parametric G-computation and Monte Carlo simulation, adjusting for confounders to simulate a controlled trial. This type of simulation is being increasingly accepted in observational research. They accounted for daily changes in patient condition (e.g., pH, SOFA score, lactate, vasopressor use). They used rolling entry matching – a type of propensity matching to add to the robustness of the comparison between groups. 

Limitations 

At the end of the day, the study carries the inherent weaknesses of a retrospective observational trial. No long-term data (beyond 30 days) were available. The use of bicarbonate was not standardized; hence, no dose-response relationship could be evaluated. The computation assumed bicarbonate was given immediately when criteria were met – this may not reflect real-world practice. The study modelled outcomes but did not assess short-term physiological responses (e.g., change in pH after bicarbonate). Although rolling entry matching also revealed reduced ICU mortality with bicarbonate, the result fell short of statistical significance (the 95% CI crossed zero)

Summary

Sodium bicarbonate is often administered to combat metabolic acidosis in critically ill patients. However, despite widespread use, there is a lack of robust evidence from controlled trials demonstrating a mortality benefit. 

This study tried to answer that question using real-world data of over 6,000 ICU patients from 12 ICUs in Australia. Instead of running a new trial, the researchers used a method called “target trial emulation”—a method of analyzing existing data to mimic what a randomized trial might have shown. They compared patients who received bicarbonate with those who did not, while carefully adjusting for baseline severity and how their condition changed day by day.

The results suggested that giving bicarbonate was linked to a small but meaningful improvement in survival, especially in patients with more severe acidosis and acute kidney injury. While not definitive, these findings support the hypothesis that bicarbonate therapy may save lives in selected patients.

Looking ahead, ongoing randomized trials will be key to confirm who benefits most—and when we should reach for the bicarb ampoule.

 

 

Best critical care blogs: https://bloggers.feedspot.com/critical_care_blogs/

References 

1.         Jung B, Rimmele T, Le Goff C, Chanques G, Corne P, Jonquet O, et al. Severe metabolic or mixed acidemia on intensive care unit admission: incidence, prognosis and administration of buffer therapy. A prospective, multiple-center study. Crit Care. 2011;15(5):R238. 

2.         Wardi G, Holgren S, Gupta A, Sobel J, Birch A, Pearce A, et al. A Review of Bicarbonate Use in Common Clinical Scenarios. J Emerg Med. 2023 Aug;65(2):e71–80. 

3.         Joannes-Boyau O, Forni LG. Time to treat metabolic acidosis in the ICU with sodium bicarbonate? Anaesth Crit Care Pain Med. 2018 Dec;37(6):493–4. 

4.         Cooper DJ, Walley KR, Wiggs BR, Russell JA. Bicarbonate does not improve hemodynamics in critically ill patients who have lactic acidosis. A prospective, controlled clinical study. Ann Intern Med. 1990 Apr 1;112(7):492–8. 

5.         Jaber S, Paugam C, Futier E, Lefrant JY, Lasocki S, Lescot T, 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 Jul 7;392(10141):31–40. 

6.         Blank SP, Blank RM, Laupland KB, Tabah A, Gill D, Kumar A, et al. Sodium bicarbonate administration for metabolic acidosis in the intensive care unit: a target trial emulation. Intensive Care Med. 2025 Jun;51(6):1–9.