1. Publication
- Title: Sodium Bicarbonate for Severe Metabolic Acidemia and Acute Kidney Injury: The BICARICU-2 Randomized Clinical Trial
- Acronym: BICARICU-2
- Year & Journal: JAMA, epublished October 29, 2025 (print: 2025;334(22):2000-2010)
- Citation: Jung B, Jabaudon M, De Jong A, et al; BICARICU-2 Study Group. JAMA. 2025;334(22):2000-2010. doi:10.1001/jama.2025.20231
2. Context & Rationale
Background: Predecessor BICAR-ICU found no overall benefit from bicarbonate, but a prespecified AKI stage 2-3 subgroup showed lower mortality (46% vs 63%) and less RRT use — a specific, testable hypothesis, further supported by an Australian target trial emulation.
Research Question/Hypothesis: In critically ill adults with severe metabolic acidemia (pH≤7.20) and moderate-severe AKI, does IV sodium bicarbonate (target pH≥7.30) reduce day-90 mortality vs no bicarbonate?
Why This Matters: Directly, appropriately tests the specific AKI-subgroup hypothesis generated by BICAR-ICU rather than repeating an all-comer design.
3. Design & Methods
- Study Type: Multicenter, open-label, pragmatic RCT
- Setting & Centers: 43 ICUs, France; 2019-2023
- Population: Severe metabolic acidemia (pH≤7.20) + KDIGO stage 2-3 AKI. Median age 67y, predominantly septic shock, ~80% vasopressors, ~75% ventilated.
- Intervention: IV sodium bicarbonate (4.2%) targeting pH≥7.30 (n=314)
- Comparator: No bicarbonate (n=313)
- Randomization: 1:1, 640 enrolled, 627 analyzed (ITT)
- Blinding: Open-label
- Statistical Power & Follow-Up: Primary: day-90 mortality. Secondary: day-28/180 mortality, organ support, RRT use to day 28.
4. Key Results
Outcome | Bicarbonate | Control | Effect Size | 95% CI | p-value | Notes |
Day-90 mortality (primary) | 62.1% | 61.7% | Diff 0.4pp | −7.2 to 8.0 | 0.91 | Clear null |
Cumulative RRT incidence by day 28 | Lower | Higher | HR 0.59 | 0.46–0.75 | — | Significant reduction |
Median time to RRT initiation | 31h | 16h | — | — | — | RRT delayed |
Acidosis-driven RRT trigger frequency | 41% | 64% | — | — | — | Fewer acidosis-triggered RRT starts |
Nuance (NephJC): Main RRT triggers (oliguria ~85%, refractory acidosis, rising urea/creatinine) were similar between groups — the RRT-reduction reflects clinician response to corrected acidemia more than a true kidney-recovery change.
5. Internal Validity Assessment
- Randomization & Allocation: Multicenter across 43 ICUs; concealment mechanics not detailed.
- Protocol Adherence & Separation: Clear pH-target titration; genuine biochemical separation.
- Blinding & Detection Bias: Open-label — authors themselves flag this as a limitation, especially for RRT-timing (clinician-judgment-dependent).
- Overall Internal Validity Conclusion: Strong — well-powered (640), high completeness (98% ITT), clean null primary, robust/coherent secondary finding; observed mortality (62%) notably lower than the 70-80% assumed in power calculation — relevant context.
6. External Validity Assessment
- Population Representativeness: Directly targets the exact AKI subgroup generated by BICAR-ICU's hypothesis.
- Overall External Validity Conclusion: Good — generalizable to similar severe-acidemia-with-AKI ICU populations.
7. Strengths & Limitations
Strengths: Directly tests a specific prior subgroup hypothesis; large, high-completeness; clean null plus coherent secondary finding; transparent authors' discussion.
Limitations: Open-label; lower-than-expected observed mortality may affect power; RRT-reduction reflects a threshold-response effect rather than demonstrated altered kidney trajectory.
8. Interpretation & Practice Impact
- Clinical Implications: Bicarbonate does not improve survival even in the specific subgroup where benefit was previously suggested — but meaningfully delays/reduces RRT.
- Mechanistic Coherence: Corrects the acidemia trigger for RRT without altering the disease trajectory driving mortality — symptomatic management, not disease modification.
- Systems-Level Takeaway: May still be valuable for limiting RRT-related risk exposure even without a survival benefit.
9. Controversies & Subsequent Evidence
Accompanying editorial ("Turning the Tide or Chasing a Myth?") directly engages the ongoing bicarbonate debate. The UK MOSAICC trial will add further data in a moderate-acidemia population.
10. Summary & Executive Takeaway
Summary: BICARICU-2 randomized 627 French ICU patients with severe acidemia + AKI 2-3 to bicarbonate or no bicarbonate. Day-90 mortality was unchanged (62.1% vs 61.7%, P=0.91), but RRT use was significantly reduced and delayed (HR 0.59).
Overall Takeaway: Bicarbonate doesn't improve survival even in the subgroup previously thought to benefit, but meaningfully delays/reduces RRT — a genuine secondary benefit that may justify selective use for RRT-avoidance rather than survival.
11. Bibliography
- Jaber S, et al. BICAR-ICU. Lancet. 2018;392(10141):31-40.
- Jung B, et al. BICARICU-2 protocol. BMJ Open. 2023;13:e073487.