Development and Validation of the Sequential Organ Failure Assessment (SOFA)-2 Score.

Area of Science:

• Critical Care Medicine
• Clinical Epidemiology
• Biostatistics

Background:

• The Sequential Organ Failure Assessment (SOFA) score, a standard for assessing organ dysfunction in critical illness, has not been updated in 30 years.
• This may limit its accuracy in reflecting current clinical practices and patient outcomes.

Purpose of the Study:

• To develop an updated SOFA score (SOFA-2) using a data-driven approach informed by expert consensus.
• To improve the description and prediction of organ dysfunction in critically ill patients across diverse settings.

Main Methods:

• A federated analysis of over 3.3 million patient encounters from 1319 ICUs in 9 countries (2014-2023).
• Data-driven score development and validation using multicenter cohorts, incorporating expert input via a modified Delphi process.
• Assessed predictive validity using the area under the receiver operating characteristic (AUROC) curve.

Main Results:

• SOFA-2 modified the original 6 organ systems with new variables and revised thresholds, improving the description of organ dysfunction (SOFA-2 AUROC, 0.79 vs. SOFA-1 AUROC, 0.77).
• The updated score maintained predictive validity when evaluated sequentially from ICU day 1 to day 7.
• Gastrointestinal and immune dysfunction scores could not be incorporated due to insufficient data and lack of content validity.

Conclusions:

• The SOFA-2 score offers an updated, validated measure of organ dysfunction in critically ill adults.
• It better reflects contemporary organ support treatments and utilizes refined thresholds for improved predictive accuracy.
• This score is relevant for diverse geographical and socioeconomic populations, enhancing critical care assessment.