Sapsii Score

SAPS-II Score Calculation

The Simplified Acute Physiology Score II (SAPS-II) is a widely used severity scoring system developed from a European/North American multicenter study. Calculated on the first day of ICU stay.

When to Use This Skill

• Hospital mortality prediction
• Severity stratification
• International benchmarking (widely used in European ICUs)
• Research cohort matching
• Quality improvement initiatives

Score Components (First 24 Hours)

Pre-computed Table

SELECT
    subject_id,
    hadm_id,
    stay_id,
    starttime,
    endtime,
    sapsii,
    sapsii_prob,  -- Predicted hospital mortality
    age_score,
    hr_score,
    sysbp_score,
    temp_score,
    pao2fio2_score,
    uo_score,
    bun_score,
    wbc_score,
    potassium_score,
    sodium_score,
    bicarbonate_score,
    bilirubin_score,
    gcs_score,
    comorbidity_score,
    admissiontype_score
FROM mimiciv_derived.sapsii;

Critical Implementation Notes

1. PaO2/FiO2 Scoring: Only scored for patients on mechanical ventilation OR CPAP/BiPAP. Non-ventilated patients get 0 points for this component.

2. Comorbidity Definitions (ICD-based):

   • AIDS: ICD-9 042-044, ICD-10 B20-B22, B24
   • Hematologic Malignancy: ICD-9 200xx-208xx, ICD-10 C81-C96
   • Metastatic Cancer: ICD-9 196x-199x, ICD-10 C77-C79, C800

3. Admission Type Classification:

   • Scheduled Surgical: Elective admission + surgical service
   • Unscheduled Surgical: Non-elective + surgical service
   • Medical: All others

4. GCS Handling: GCS < 3 is treated as null (erroneous or tracheostomy). Sedated patients should use pre-sedation GCS.

5. Mortality Probability: Calculated using:

   sapsii_prob = 1 / (1 + exp(-(-7.7631 + 0.0737*sapsii + 0.9971*ln(sapsii+1))))
   

6. Time Window: Uses data from ICU admission to 24 hours after.

Example: Severity Distribution

SELECT
    CASE
        WHEN sapsii < 25 THEN 'Low (<25)'
        WHEN sapsii < 50 THEN 'Moderate (25-49)'
        WHEN sapsii < 75 THEN 'High (50-74)'
        ELSE 'Very High (>=75)'
    END AS severity_category,
    COUNT(*) AS n_patients,
    ROUND(AVG(sapsii_prob), 3) AS avg_predicted_mortality
FROM mimiciv_derived.sapsii
GROUP BY 1
ORDER BY 1;

Example: Component Analysis

-- Which components contribute most to high SAPS-II?
SELECT
    'Age' AS component, AVG(age_score) AS avg_score FROM mimiciv_derived.sapsii WHERE sapsii >= 50
UNION ALL
SELECT 'Heart Rate', AVG(hr_score) FROM mimiciv_derived.sapsii WHERE sapsii >= 50
UNION ALL
SELECT 'Systolic BP', AVG(sysbp_score) FROM mimiciv_derived.sapsii WHERE sapsii >= 50
UNION ALL
SELECT 'GCS', AVG(gcs_score) FROM mimiciv_derived.sapsii WHERE sapsii >= 50
UNION ALL
SELECT 'Comorbidity', AVG(comorbidity_score) FROM mimiciv_derived.sapsii WHERE sapsii >= 50
ORDER BY avg_score DESC;

References

• Le Gall JR, Lemeshow S, Saulnier F. "A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study." JAMA. 1993;270(24):2957-2963.