Publication

• Title: Sodium bicarbonate for severe metabolic acidemia and acute kidney injury: the BICARICU-2 randomized clinical trial
• Acronym: BICARICU-2
• Year: 2025
• Journal published in: JAMA
• Citation: Jung B, Jabaudon M, De Jong A, et al; BICARICU-2 Study Group. Sodium bicarbonate for severe metabolic acidemia and acute kidney injury: the BICARICU-2 randomized clinical trial. JAMA. 2025;334(22):2000-2010.

Context & Rationale

• Background

  • Severe metabolic acidaemia in critical illness is associated with haemodynamic compromise, impaired catecholamine responsiveness, arrhythmias, and high mortality.
  • Sodium bicarbonate is widely used to buffer severe acidaemia, but it can increase CO₂ generation, sodium load, fluid load, and precipitate alkalosis/hypocalcaemia; equipoise persists regarding patient selection and net clinical benefit.
  • The earlier BICAR-ICU trial in severe metabolic acidosis was neutral overall but suggested benefit in the prespecified subgroup with acute kidney injury (AKI) stage 2–3, including less kidney replacement therapy (KRT) use.¹
  • Subsequent expert guidance endorsed bicarbonate as reasonable in severe metabolic acidosis with AKI, while acknowledging limited certainty and the need for definitive trials.²
  • Prior non-randomised evidence (including a target-trial emulation) and a small double-blind pilot RCT suggested potential benefit in selected populations, but residual confounding and imprecision limited inference.³⁴
• Research Question/Hypothesis

  • Does protocolised sodium bicarbonate, titrated to achieve and maintain arterial pH ≥7.30, reduce 90-day all-cause mortality in ICU patients with severe metabolic acidaemia and AKI stage 2–3 compared with usual care?
  • Key mechanistic corollary: does bicarbonate reduce KRT use (and major adverse kidney events) without causing clinically important harm?
• Why This Matters

  • This phenotype carries high risk (observed ~62% 90-day mortality) and frequent KRT exposure (~50% in usual care), making even modest improvements clinically meaningful.
  • Clarifying whether bicarbonate improves patient-centred outcomes (mortality, renal recovery) would standardise a common but variably applied intervention.
  • Defining the safety trade-offs (alkalosis, hypernatraemia, hypocalcaemia, fluid balance) is essential before recommending routine use.

Design & Methods

• Research Question: In critically ill adults with severe metabolic acidaemia and AKI stage 2–3, does IV sodium bicarbonate (4.2%) targeting arterial pH ≥7.30 reduce 90-day mortality compared with usual care?
• Study Type: Investigator-initiated, multicentre (43 French ICUs), parallel-group, randomised trial using minimisation/stratification by centre, age (<65 vs ≥65 years), and pH (<7.10 vs 7.11–7.20); open-label (no placebo). Enrolment October 2019 to December 2023; last follow-up June 2024.
• Population:
  • Inclusion: ICU adults with AKI stage 2 or 3 (KDIGO) and severe metabolic acidaemia defined by ≥2 arterial blood gases ≥30 minutes apart with pH ≤7.20 and bicarbonate ≤20 mEq/L, with PaCO₂ ≤45 mm Hg.
  • Additional severity requirement: total SOFA score >4 and/or arterial lactate ≥2 mmol/L within 48 hours after ICU admission.
  • Key exclusions: immediate indication for KRT (eg, K >6.5 mmol/L with ECG changes and/or cardiogenic pulmonary oedema with anuria and PaO₂/FiO₂ <200 while FiO₂ >50% and PEEP >5 cm H₂O); primary respiratory acidaemia (PaCO₂ >45 mm Hg); diabetic ketoacidosis; acute gastrointestinal bicarbonate losses.
• Intervention:
  • IV sodium bicarbonate 4.2% (0.5 mmol/mL), administered in 125–250 mL infusions over 30 minutes, repeated as needed to reach and maintain arterial pH ≥7.30.
  • Maximum total bicarbonate volume 1000 mL in the first 24 hours; subsequent administration permitted up to day 28 if pH fell below target.
  • KRT initiation: immediate indications as above; otherwise KRT was recommended if ≥2 criteria were present ≥24 hours after enrolment, including urine output <0.3 mL/kg/h for 24 hours, K >6.5 mmol/L, and/or pH <7.20 despite bicarbonate administration for ≥6 hours.
• Comparison:
  • Usual care without protocolised bicarbonate administration.
  • KRT initiation: immediate indications as above; otherwise KRT was recommended if ≥2 criteria were present ≥24 hours after enrolment, including urine output <0.3 mL/kg/h for 24 hours, K >6.5 mmol/L, and/or pH ≤7.20 despite resuscitation.
• Blinding: Unblinded (open-label). Outcomes such as mortality are objective; KRT initiation and some secondary outcomes are potentially vulnerable to performance/decision bias.
• Statistics: Power: 588 patients were required to detect a 10% absolute reduction in 90-day mortality (from 80% to 70%) with 80% power at the 5% (two-sided) significance level, with interim monitoring (Haybittle–Peto boundary P<0.001); planned enrolment was 640 allowing ~8% unevaluable. Primary analysis: modified intention-to-treat (excluded prespecified post-randomisation ineligible/outlier patients); effect estimates reported as absolute differences with 95% CIs; time-to-event outcomes used Cox models; secondary outcomes interpreted as exploratory (no multiplicity adjustment).⁵
• Follow-Up Period: 180 days (primary endpoint at 90 days).

Key Results

This trial was not stopped early. It completed planned enrolment (640 randomised) and follow-up (primary outcome at 90 days).

Outcome	Sodium bicarbonate	Usual care	Effect	p value / 95% CI	Notes
Day 90 all-cause mortality (primary)	195/314 (62.1%)	193/313 (61.7%)	Absolute difference 0.4%	95% CI -7.2 to 8.0; P=0.91	Primary endpoint; modified intention-to-treat population (n=627)
Day 180 all-cause mortality	217/314 (69.1%)	218/313 (69.6%)	Absolute difference -0.5%	95% CI -7.7 to 6.7; P not reported	Exploratory secondary outcome
Any kidney replacement therapy	109/314 (34.7%)	157/313 (50.2%)	Absolute difference -15.5%	95% CI -23.1 to -7.8; HR 0.59; 95% CI 0.46 to 0.75; P<0.001	Secondary outcome; KRT initiation recommended per prespecified criteria
Time to KRT initiation (among those who received KRT)	31 h (12–70)	18 h (5–37)	Median difference 10 h	95% CI 7 to 20; P not reported	Hours from enrolment to KRT start
Major adverse kidney events by day 90	188/314 (59.9%)	215/313 (68.7%)	Absolute difference -8.8%	95% CI -16.3 to -1.3; P not reported	Composite kidney outcome; exploratory
ICU-acquired bloodstream infection	14/314 (4.5%)	28/313 (8.9%)	Absolute difference -4.5%	95% CI -8.4 to -0.6; P not reported	Exploratory secondary outcome
Fluid balance at 48 h (H0–H48), mL	2805 (750–5950)	2220 (0–4870)	Median difference 954 mL	95% CI 320 to 1615; P not reported	Protocolised bicarbonate increased early net fluid balance
Blood pH >7.50 (≥1 episode)	87/314 (28%)	73/313 (23%)	Absolute difference 4.4%	95% CI -2.4 to 11.2; P not reported	Adverse metabolic event
Serum sodium >155 mmol/L (≥1 episode)	15/314 (4.8%)	7/313 (2.2%)	Absolute difference 2.5%	95% CI -0.3 to 5.4; P not reported	Adverse metabolic event

• Protocolised bicarbonate did not reduce 90-day mortality (62.1% vs 61.7%; absolute difference 0.4%; 95% CI -7.2 to 8.0; P=0.91).
• Bicarbonate was associated with substantially less KRT use (34.7% vs 50.2%; absolute difference -15.5%; 95% CI -23.1 to -7.8) and delayed KRT initiation (median 31 h vs 18 h among those treated).
• Early net fluid balance was higher with bicarbonate at 48 hours (median 2805 mL vs 2220 mL; median difference 954 mL; 95% CI 320 to 1615); metabolic alkalosis/hypernatraemia events were uncommon and CIs included both harm and no harm.

Internal Validity

• Randomisation and Allocation: Central randomisation using minimisation/stratification (centre, age, pH stratum), supporting baseline balance and allocation concealment up to randomisation.
• Drop out or exclusions: 640 randomised; 627 analysed (314 bicarbonate, 313 usual care). Post-randomisation exclusions included ineligibility/outlier physiology (eg, pH >7.24, bicarbonate >20 mEq/L, PaCO₂ >47 mm Hg) and not meeting KDIGO stage 2–3.
• Performance/Detection Bias: Open-label design with no placebo; mortality is objective, but clinician-driven decisions (notably KRT initiation and some secondary outcomes) are susceptible to performance bias.
• Protocol Adherence: Bicarbonate administered to 296/314 (94.3%) in the intervention group; bicarbonate administered to 47/313 (15.0%) in the usual-care group (contamination).
• Baseline Characteristics: Groups were similar at baseline (eg, median age 67 vs 68 years; median pH 7.15 vs 7.15; median lactate 5.8 vs 5.9 mmol/L; septic shock 55% vs 53%).
• Heterogeneity: Conducted across 43 ICUs; prespecified subgroup analyses did not demonstrate clear effect modification for the primary outcome (eg, pH <7.10: OR 0.62; 95% CI 0.34 to 1.15; pH 7.11–7.20: OR 1.23; 95% CI 0.84 to 1.81).
• Timing: Intervention delivered early after enrolment (median 15 minutes [IQR 6–36]) and allowed up to day 28 if acidaemia recurred.
• Dose: Protocol cap of 1000 mL bicarbonate in the first 24 hours; delivered exposure suggests substantial early dosing (H0–H48 bicarbonate volume 750 mL [500–1000] vs 0 [0–0]).
• Separation of the Variable of Interest: H0–H48 bicarbonate volume 750 mL (500–1000) vs 0 mL (0–0); H0–H48 net fluid balance 2805 mL (750–5950) vs 2220 mL (0–4870); crystalloid volume H0–H48 3215 mL (1500–6000) vs 3000 mL (1500–5000).
• Crossover: Contamination occurred (47/313 usual-care patients received bicarbonate). A per-protocol analysis (exposure-defined) reported day-90 mortality 222/349 (64%) with bicarbonate exposure vs 148/252 (59%) without; absolute difference 4.9% (95% CI -3.0 to 12.8) and KRT use 132/349 (38%) vs 119/252 (47%); absolute difference -8.4% (95% CI -16.4 to -0.4).
• Outcome Assessment: Mortality is objective; KRT initiation was guided by prespecified criteria but remains clinician-implemented; infections were prespecified secondary outcomes.
• Statistical Rigor: Prespecified interim monitoring (P<0.001); analyses included unadjusted and adjusted models for primary outcome; secondary outcomes were exploratory without multiplicity adjustment.

Conclusion on Internal Validity: Moderate. Randomisation and follow-up were robust with good separation in bicarbonate exposure, but the open-label design, contamination, post-randomisation exclusions, and decision-sensitive secondary endpoints (notably KRT) temper causal certainty for non-mortality outcomes.

External Validity

• Population Representativeness: Typical high-acuity ICU population with severe metabolic acidaemia and substantial shock burden; exclusions (hypercapnic acidaemia, diabetic ketoacidosis, acute GI bicarbonate loss, and immediate KRT indications) narrow applicability to specific acidaemia phenotypes.
• Applicability: Likely generalisable to similarly resourced ICUs with ready access to arterial blood gas monitoring and KRT; generalisability may be limited where bicarbonate concentration/availability differs (4.2% was used) or where KRT pathways and thresholds differ.

Conclusion on External Validity: Generalisability is moderate for adult ICUs managing severe metabolic acidaemia with AKI stage 2–3, but limited for hypercapnic acidaemia, ketoacidosis/toxin-related acidaemia, and patients requiring immediate KRT.

Strengths & Limitations

• Strengths: Large, multicentre randomised trial in a high-risk population; pragmatic protocol reflecting real ICU practice; explicit KRT decision framework; excellent follow-up for mortality; clinically relevant outcomes (mortality, KRT, MAKE).
• Limitations: Open-label without placebo; contamination in the usual-care group; KRT initiation is clinician-implemented and partly dependent on pH (a modifiable treatment target); post-randomisation exclusions; trial powered for a larger mortality effect than ultimately observed event rates; secondary outcomes not adjusted for multiplicity.

Interpretation & Why It Matters

• Mortality

  Protocolised bicarbonate targeting pH ≥7.30 did not improve 90-day (or 180-day) survival in severe metabolic acidaemia with AKI stage 2–3.
• Kidney-related outcomes

  Despite no mortality signal, bicarbonate was associated with fewer patients receiving KRT and fewer major adverse kidney events by day 90, suggesting potential value in reducing KRT exposure and its downstream complications.
• Safety and implementation

  Clinicians should anticipate increased early fluid balance and monitor for alkalosis and hypernatraemia; severe electrolyte derangements were uncommon but not absent.

Controversies & Subsequent Evidence

• Power and effect size assumptions: The trial was powered for a 10% absolute mortality reduction with an assumed 80% control mortality; observed 90-day mortality (~62%) implies reduced power for smaller, plausible effects.
• Decision-sensitive endpoints in an unblinded trial: KRT initiation criteria included pH thresholds, a direct treatment target, raising interpretive ambiguity about whether reduced KRT reflects renal recovery or avoidance/delay of meeting the KRT trigger; the accompanying editorial emphasised this limitation for mechanistic interpretation of kidney endpoints.⁶
• Minimisation randomisation analysis considerations: Trials using minimisation/stratified allocation can require appropriate adjusted analyses for valid inference; BICARICU-2 reported both unadjusted and adjusted analyses with similar conclusions.⁷
• Contamination: 15% of usual-care patients received bicarbonate, which plausibly diluted differences in both mortality and pH-driven downstream decisions.
• Relationship to KRT timing evidence: Large RCTs comparing early vs delayed KRT initiation in critical illness generally show reduced KRT exposure without clear mortality benefit, contextualising BICARICU-2’s “less KRT without survival benefit” signal as biologically and clinically plausible.⁸⁹¹⁰
• Guideline and practice implications: Contemporary sepsis guidance continues to prioritise resuscitation and treatment of the cause of shock and lactic acidaemia; bicarbonate remains selectively considered rather than routine for mortality benefit, with ongoing uncertainty about the “right” phenotype and thresholds for buffering therapy.¹¹
• Subsequent and adjacent evidence: A target-trial emulation suggested benefit in severe metabolic acidaemia, whereas randomised evidence (including a pilot double-blind trial) remains imprecise and now includes a large null mortality RCT (BICARICU-2), reinforcing the need to prioritise patient-centred outcomes over pH correction alone.³⁴

Summary

• BICARICU-2 randomised 640 ICU patients with severe metabolic acidaemia (pH ≤7.20; bicarbonate ≤20) and AKI stage 2–3; 627 were included in the primary analysis.
• Protocolised sodium bicarbonate (4.2%) targeting pH ≥7.30 did not reduce 90-day mortality (62.1% vs 61.7%).
• Bicarbonate was associated with fewer patients receiving KRT (34.7% vs 50.2%) and fewer major adverse kidney events by day 90 (59.9% vs 68.7%).
• Early net fluid balance was higher with bicarbonate; metabolic alkalosis and hypernatraemia occurred but were infrequent.
• The open-label design, contamination, and pH-dependent KRT criteria are central to interpreting kidney-related secondary outcomes.

Further Reading

Other Trials

• 2018Jaber S, Paugam-Burtz C, Dupuy AM, et al. Sodium bicarbonate therapy for patients with severe metabolic acidosis in the intensive care unit (BICAR-ICU): a multicentre, open-label, randomised controlled, phase 3 trial. Lancet. 2018;392(10141):31-40.
• 2023Serpa Neto A, et al. Sodium bicarbonate vs placebo in patients with metabolic acidosis and acute kidney injury: a double-blind pilot randomised clinical trial. Crit Care Med. 2023;51(10):1478-1488.
• 2016Gaudry S, Hajage D, Schortgen F, et al. Initiation strategies for renal-replacement therapy in the intensive care unit. N Engl J Med. 2016;375(2):122-133.
• 2018Barbar SD, Clere-Jehl R, Bourredjem A, et al. Timing of renal-replacement therapy in patients with acute kidney injury and sepsis. N Engl J Med. 2018;379(15):1431-1442.
• 2020STARRT-AKI Investigators. Timing of initiation of renal-replacement therapy in acute kidney injury. N Engl J Med. 2020;383(3):240-251.

Systematic Review & Meta Analysis

• 2025Cheng D, et al. Effect of sodium bicarbonate on mortality in critically ill patients with metabolic acidosis: a systematic review and meta-analysis. PLoS One. 2025;20(4):e0324732.
• 2024Zhou Y, et al. Sodium bicarbonate therapy in critically ill patients with metabolic acidosis: systematic review and meta-analysis. Crit Care. 2024;28(1):386.
• 2024Wang Y, et al. Sodium bicarbonate administration and outcomes in critically ill patients with metabolic acidosis: systematic review and meta-analysis. BMC Nephrol. 2024;25(1):196.
• 2023Alshahrani MS, et al. Sodium bicarbonate in critically ill patients with metabolic acidosis: systematic review and meta-analysis. BMJ Open. 2023;13(12):e074199.
• 2022Zhang Z, et al. Effects of sodium bicarbonate on mortality in patients with metabolic acidosis: systematic review and meta-analysis. Front Pharmacol. 2022;13:759247.

Observational Studies

• 2011Jung B, Rimmele T, Le Goff C, et al. Severe metabolic or mixed acidaemia on ICU admission: incidence, prognosis and administration of buffer therapy. Crit Care. 2011;15(5):R238.
• 2018Zhang Z, Zhu C, Mo L, Hong Y. Effectiveness of sodium bicarbonate infusion on mortality in septic patients with metabolic acidosis. Intensive Care Med. 2018;44(11):1888-1895.
• 2019Zhang Z, Mo L, Ho KM, Hong Y. Association between use of sodium bicarbonate and mortality in acute kidney injury in patients with metabolic acidosis: a marginal structural Cox model. Crit Care Med. 2019;47(10):1402-1408.
• 2023Yan M, Peng Z, Zhang X, et al. Association between sodium bicarbonate use and mortality in moderate metabolic acidosis in the ICU: a retrospective cohort study. Front Med. 2023;10:1268252.
• 2025Blank M, Hannemann A, Robinson K, et al. Management of severe metabolic acidosis with sodium bicarbonate in the ICU: a target trial emulation. Intensive Care Med. 2025;51(5):733-743.

Guidelines

• 2021Evans L, Rhodes A, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock 2021. Crit Care Med. 2021;49(11):e1063-e1143.
• 2017Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock: 2016. Intensive Care Med. 2017;43(3):304-377.
• 2019Jung B, Martinez M, Claessens YE, et al. Diagnosis and management of metabolic acidosis: guidelines from a French expert panel. Ann Intensive Care. 2019;9(1):92.
• 2022Wald R, Beaubien-Souligny W, Chanchlani R, et al. Delivering optimal renal replacement therapy to critically ill patients with acute kidney injury. Intensive Care Med. 2022;48(10):1368-1381.
• 2023Ostermann M, Bagshaw SM, Lumlertgul N, Wald R. Indications for and timing of initiation of kidney replacement therapy. Clin J Am Soc Nephrol. 2023;18(1):113-120.

Notes

• When translating BICARICU-2 to practice, interpret the reduction in KRT alongside the pH-dependent KRT criteria and open-label design; prioritise patient-centred outcomes and local KRT pathways.

Overall Takeaway

BICARICU-2 is a definitive modern trial testing protocolised bicarbonate in a high-risk ICU phenotype (severe metabolic acidaemia plus AKI stage 2–3). It shows no survival benefit, but does show fewer patients receiving KRT and fewer major adverse kidney events, at the cost of greater early fluid balance and modest metabolic adverse events. The trial reshapes practice by separating “pH correction” from “mortality benefit” while sharpening the debate about kidney endpoints when treatment directly modifies KRT triggers.

Bibliography

• 1.Jaber S, Paugam-Burtz C, Dupuy AM, et al. Sodium bicarbonate therapy for patients with severe metabolic acidosis in the intensive care unit (BICAR-ICU): a multicentre, open-label, randomised controlled, phase 3 trial. Lancet. 2018;392(10141):31-40.
• 2.Jung B, Martinez M, Claessens YE, et al. Diagnosis and management of metabolic acidosis: guidelines from a French expert panel. Ann Intensive Care. 2019;9(1):92.
• 3.Blank M, Hannemann A, Robinson K, et al. Management of severe metabolic acidosis with sodium bicarbonate in the ICU: a target trial emulation. Intensive Care Med. 2025;51(5):733-743.
• 4.Serpa Neto A, et al. Sodium bicarbonate vs placebo in patients with metabolic acidosis and acute kidney injury: a double-blind pilot randomised clinical trial. Crit Care Med. 2023;51(10):1478-1488.
• 5.Jung B, Chartier JP, De Jong A, et al. Sodium bicarbonate therapy for patients with severe metabolic acidosis and acute kidney injury in the intensive care unit: protocol for the BICARICU-2 randomised clinical trial. BMJ Open. 2023;13:e073487.
• 6.Kusirisin P, Zampieri FG, Bagshaw SM. Sodium bicarbonate in severe acidemia and acute kidney injury—turning the tide or chasing a myth? JAMA. Published online October 29, 2025.
• 7.Kahan BC, Morris TP. Improper analysis of trials randomised using stratified blocks or minimisation. Stat Med. 2012;31(4):328-340.
• 8.Gaudry S, Hajage D, Schortgen F, et al. Initiation strategies for renal-replacement therapy in the intensive care unit. N Engl J Med. 2016;375(2):122-133.
• 9.Barbar SD, Clere-Jehl R, Bourredjem A, et al. Timing of renal-replacement therapy in patients with acute kidney injury and sepsis. N Engl J Med. 2018;379(15):1431-1442.
• 10.STARRT-AKI Investigators. Timing of initiation of renal-replacement therapy in acute kidney injury. N Engl J Med. 2020;383(3):240-251.
• 11.Evans L, Rhodes A, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock 2021. Crit Care Med. 2021;49(11):e1063-e1143.