Section 1.1: What’s New in the Sequential Organ Failure Assessment (SOFA)-2 Score?
Multi-organ dysfunction syndrome (MODS) remains a leading cause of morbidity and mortality in critically ill patients. The evaluation of organ dysfunction and/or failure is typically guided by three core principles:
Organ dysfunction is a continuous, quantifiable process rather than a binary “all-or-nothing” event.
Organ function in critically ill patients changes rapidly, necessitating at least daily reassessment to accurately track disease progression.
Assessment metrics should be as simple, accessible, and reproducible as possible.
In 1996, the European Society of Intensive Care Medicine (ESICM) Working Group introduced the Sepsis-Related Organ Failure Assessment (SOFA) score, which was renamed the Sequential Organ Failure Assessment (SOFA) score in 1998. The SOFA score evaluates six organ systems (respiratory, coagulation, liver, cardiovascular, central nervous system, and renal) independently, with the cumulative score reflecting the overall severity of organ dysfunction. Unlike the APACHE score, which predicts hospital mortality based on initial 24-hour physiological parameters, SOFA was originally designed to track organ failure trajectories rather than predict outcomes.
| Organ System | 0 Points | 1 Point | 2 Points | 3 Points | 4 Points |
|---|---|---|---|---|---|
| Respiratory | PaO₂/FiO₂ ≥ 400 mmHg | PaO₂/FiO₂ < 400 mmHg | PaO₂/FiO₂ < 300 mmHg | PaO₂/FiO₂ < 200 mmHg (on mechanical ventilation) | PaO₂/FiO₂ < 100 mmHg (on mechanical ventilation) |
| Coagulation | Platelets ≥ 150 × 10⁹/L | Platelets < 150 × 10⁹/L | Platelets < 100 × 10⁹/L | Platelets < 50 × 10⁹/L | Platelets < 20 × 10⁹/L |
| Liver | Total bilirubin < 20 μmol/L | 20–32 μmol/L | 33–101 μmol/L | 102–204 μmol/L | > 204 μmol/L |
| Cardiovascular | MAP ≥ 70 mmHg | MAP < 70 mmHg | Dopamine ≤ 5 μg/kg/min or any dose of dobutamine | Dopamine > 5 μg/kg/min or Epinephrine ≤ 0.1 μg/kg/min or Norepinephrine ≤ 0.1 μg/kg/min | Dopamine > 15 μg/kg/min or Epinephrine > 0.1 μg/kg/min or Norepinephrine > 0.1 μg/kg/min |
| Central Nervous System | GCS 15 | GCS 13–14 | GCS 10–12 | GCS 6–9 | GCS < 6 |
| Renal | Creatinine < 110 μmol/L | 110–170 μmol/L | 171–299 μmol/L | 300–440 μmol/L or Urine output < 500 mL/d | > 440 μmol/L or Urine output < 200 mL/d |
Note: PaO₂ = arterial partial pressure of oxygen; FiO₂ = fraction of inspired oxygen; GCS = Glasgow Coma Scale; MAP = mean arterial pressure.
I. Rationale for SOFA Score Revision
Over the past three decades, the SOFA score has become ubiquitous in clinical practice and research, including sepsis diagnosis. However, its limitations have become increasingly apparent:
How should GCS be scored in sedated or paralyzed patients?
How are alternative vasopressors (vasopressin, angiotensin II, phenylephrine) and inotropes (levosimendan, PDE-3 inhibitors) incorporated?
When is SpO₂/FiO₂ an acceptable substitute for PaO₂/FiO₂?
How do organ support technologies (RRT, ECMO, IABP, LVAD) impact scoring?
To address these gaps, a global consortium of 60 experts conducted a 3-year revision process (March 2022–February 2025), resulting in the SOFA-2 score.
II. Key Revisions in SOFA-2
Predictive Validity: Unlike SOFA-1, which was designed to describe dysfunction severity, SOFA-2 explicitly incorporates predictive validity (ICU mortality) into its content validity framework.
Data-Driven Thresholds: SOFA-1 thresholds were expert-derived. SOFA-2 thresholds were statistically optimized using ~2.1 million ICU patients across Australia, New Zealand, Austria, the US, and Brazil, with external validation in >1.24 million patients across six additional countries.
Terminology Updates: Organ systems were renamed for clinical precision: Central Nervous System → Brain, Coagulation → Hemostasis, Renal → Kidney, Hepatic → Liver.
Revised Scoring Thresholds: Statistical modeling adjusted several cutoffs. For example, GCS 9 scores 2 points in SOFA-2 (vs. 3 in SOFA-1). A PaO₂/FiO₂ of 250 mmHg scores 1 point (vs. 2 in SOFA-1). Creatinine, bilirubin, and platelet thresholds were similarly refined.
New Clinical Indices: Delirium treatment now scores 1 point even if GCS is 15. Cardiovascular scoring now includes all vasopressors/inotropes and mechanical circulatory support.
Organ Support Integration: ECMO (all modes) scores 4 points for respiratory support; if used for circulatory support, it scores 4 points for both cardiovascular and respiratory systems. IABP, LVAD, and microaxial pumps score 4 points for cardiovascular function. RRT scores 4 points for renal function.
Missing Data Imputation: Day 1 missing values default to 0 (assumed normal). Subsequent missing values carry forward the last recorded score (assumed stable).
Alternative Metrics: SpO₂/FiO₂ may substitute for PaO₂/FiO₂ when arterial blood gas is unavailable (SpO₂ < 98%). If full GCS assessment is impossible, the best motor response may be used.
Norepinephrine Dosing Clarification: SOFA-2 explicitly addresses the norepinephrine bitartrate salt vs. base discrepancy. Clinically administered norepinephrine is typically bitartrate (MW 337.28), meaning 2 mg of salt ≈ 1 mg of base. SOFA-2 mandates dosing be expressed in base equivalents to prevent severity overestimation.
| Organ System | 0 Points | 1 Point | 2 Points | 3 Points | 4 Points |
|---|---|---|---|---|---|
| Respiratory | PaO₂/FiO₂ > 300 mmHg | PaO₂/FiO₂ ≤ 300 mmHg | PaO₂/FiO₂ ≤ 225 mmHg | PaO₂/FiO₂ ≤ 150 mmHg + advanced ventilation support | PaO₂/FiO₂ ≤ 75 mmHg + advanced ventilation support or ECMO |
| Hemostasis | Platelets > 150 × 10⁹/L | Platelets ≤ 150 × 10⁹/L | Platelets ≤ 100 × 10⁹/L | Platelets ≤ 80 × 10⁹/L | Platelets ≤ 50 × 10⁹/L |
| Liver | Total bilirubin ≤ 20.6 μmol/L | ≤ 51.3 μmol/L | ≤ 102.6 μmol/L | ≤ 205 μmol/L | > 205 μmol/L |
| Cardiovascular | MAP ≥ 70 mmHg, no vasoactive/inotropic drugs | MAP < 70 mmHg OR low-dose NE+Epi ≤ 0.2 μg/kg/min | Medium-dose NE+Epi 0.2–0.4 μg/kg/min | High-dose NE+Epi > 0.4 μg/kg/min OR any other vasopressor/inotrope | Any mechanical circulatory support (VA-ECMO, IABP, LVAD, etc.) |
| Brain | GCS 15 (or obeys commands) | GCS 13–14 (or localizes pain) OR on delirium medication | GCS 9–12 (or withdraws to pain) | GCS 6–8 (or abnormal flexion) | GCS 3–5 (or extension/none, generalized myoclonus) |
| Kidney | Creatinine ≤ 110 μmol/L | ≤ 170 μmol/L | ≤ 300 μmol/L or UO < 0.5 mL/kg/h for ≥12h | > 300 μmol/L or UO < 0.3 mL/kg/h for ≥24h or anuria | On RRT or meets RRT criteria (Cr > 110, oliguria >6h + K⁺ ≥6.0, or pH ≤7.20 + HCO₃⁻ ≤12) |
Note: Advanced ventilation support includes HFNC, CPAP, BiPAP, NIV, invasive mechanical ventilation, or long-term home ventilation. ECMO for respiratory failure scores 4 in respiratory SOFA; for circulatory failure, it scores 4 in both cardiovascular and respiratory SOFA. Norepinephrine doses must be expressed as base equivalents. RRT criteria exclude non-renal indications (e.g., toxin clearance).
| Organ System | Changes in SOFA-2 vs. SOFA-1 |
|---|---|
| Brain | • Added delirium treatment as a scoring criterion (scores 1 point even with GCS 15) • Revised GCS thresholds • Renamed from Central Nervous System to Brain |
| Respiratory | • Incorporated non-invasive/advanced ventilation modes and ECMO • Revised PaO₂/FiO₂ thresholds |
| Cardiovascular | • Summation of catecholamine equivalents • Inclusion of all vasopressors, inotropes, and mechanical circulatory support |
| Liver | • Revised bilirubin thresholds |
| Kidney | • Revised creatinine and urine output thresholds • Explicit inclusion of RRT or meeting RRT criteria |
| Hemostasis | • Revised platelet thresholds • Renamed from Coagulation to Hemostasis |
| General Principles | • Clear definitions for each organ dysfunction level • Alternative metrics when primary data unavailable or treatment restricted • Explicit handling of sedation and chronic organ dysfunction • Standardized missing data imputation (Day 1 = 0; subsequent = last known) |
III. Clinical Implications & Limitations
SOFA-2 remains limited to six organ systems. Attempts to incorporate gastrointestinal and immune function were abandoned due to metric complexity and lack of validated biomarkers. Additionally, while SOFA-2 scores 0–4 per system, mortality risk per point varies significantly by organ (e.g., cardiovascular score 4 ≈ 50% ICU mortality vs. renal score 4 ≈ 20%). Chronic organ failure exacerbations (beyond maintenance dialysis) also lack standardized scoring adjustments.
Despite these limitations, SOFA-2’s data-driven thresholds, explicit organ support integration, and standardized missing-data rules significantly enhance reproducibility and clinical utility. Its adoption will likely reshape sepsis phenotyping, trial stratification, and bedside monitoring paradigms.
(Authors: Department of Critical Care Medicine, Peking Union Medical College Hospital)
References
Moreno R, Rhodes A, Piquilloud L, et al. The sequential organ failure assessment (SOFA) score: has the time come for an update? [J]. Crit Care, 2023, 27(1): 15.
Vincent JL, Moreno R, Takala J, et al. The SOFA (sepsis-related organ failure assessment) score to describe organ dysfunction/failure: on behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine [J]. Intensive Care Med, 1996, 22(7): 707-710.
Vincent JL, de Mendonça A, Cantraine F, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Working group on “sepsis-related problems” of the European Society of Intensive Care Medicine [J]. Crit Care Med, 1998, 26(11): 1793-1800.
Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions for sepsis and septic shock (Sepsis-3) [J]. JAMA, 2016, 315(8): 801-810.
Ranzani OT, Singer M, Salluh JIF, et al. Development and validation of the sequential organ failure assessment (SOFA)-2 score [J]. JAMA, 2025, 334(23): 2090-2103.
Moreno R, Rhodes A, Ranzani O, et al. Rationale and methodological approach underlying the development of the sequential organ failure assessment (SOFA)-2 Score: a consensus statement [J]. JAMA Netw Open, 2025, 8(10): e2545040.
Leone M, Goyer I, Levy B, et al. Dose of norepinephrine: the devil is in the details [J]. Intensive Care Med, 2022, 48(5): 638-640.
Wieruszewski PM, Leone M, Kasshanen BS, et al. Position paper on the reporting of norepinephrine formulations in critical care from the Society of Critical Care Medicine and European Society of Intensive Care Medicine Joint Task Force [J]. Crit Care Med, 2024, 52(4): 521-530.
Seymour CW. A revision to organ failure assessment in critically ill patients [J]. JAMA, 2025, 334(23): 2075-2077.