What is the APACHE II score?

APACHE II (Acute Physiology and Chronic Health Evaluation II) is an ICU severity scoring system that uses 12 physiologic variables (worst values in the first 24 hours), age points, and chronic health points to predict hospital mortality. Scores range from 0 to 71.

How is APACHE II used clinically?

APACHE II is used for ICU benchmarking, clinical trial stratification, goals-of-care discussions, and inter-ICU quality comparisons. It is not intended as the sole determinant of treatment decisions for individual patients. A standardized mortality ratio (SMR) calculated from APACHE II helps assess ICU performance.

What values should I use for APACHE II?

Use the worst (most abnormal) values recorded during the first 24 hours of ICU admission. If a lab test was not performed, use normal values. The [GCS](/tools/glasgow-coma-scale) should be assessed before sedation when possible, or document baseline as 15 if pre-sedation GCS is unavailable.

What are the chronic health conditions for APACHE II?

Chronic health points (2 or 5) are given for: cirrhosis, portal hypertension, hepatic failure, lymphoma, metastatic cancer, leukemia, AIDS, or immunosuppression from steroids or chemotherapy. Elective post-op patients get 2 points; emergency/non-operative patients get 5 points.

What APACHE II score predicts high mortality?

Score 0–9: <10% mortality. Score 10–19: 15–25%. Score 20–29: 40–55%. Score ≥30: 55–75%+. These are population estimates; individual patient outcomes vary substantially based on diagnosis, trajectory, and quality of care.

How does APACHE II differ from SOFA score?

APACHE II is calculated once at ICU admission for severity scoring and benchmarking. [SOFA Score](/tools/sofa-score) is calculated daily to track organ failure progression. SOFA is preferred for Sepsis-3 diagnosis. Both tools are complementary — APACHE II for admission severity, SOFA for daily monitoring.

Does APACHE II apply to pediatric patients?

No. APACHE II was validated only in adult ICU patients (≥16 years). Pediatric ICU patients require age-specific tools such as PIM3 or PELOD-2. Do not apply APACHE II to children.

Why is the creatinine score doubled in acute renal failure?

Acute kidney injury has a different prognostic significance than chronic kidney disease. The APACHE II model doubles the creatinine score if acute renal failure is present, reflecting the higher mortality associated with sudden-onset renal dysfunction compared to stable chronic renal impairment.

Is APACHE II still accurate with modern ICU care?

APACHE II was trained on 1980s data (Knaus et al., 1985). Modern ICU advances mean actual mortality is often lower than APACHE II predicts. Many high-performing ICUs have SMR <1. APACHE IV (2006) and SAPS III have improved calibration for modern populations, but APACHE II remains widely used due to familiarity and extensive benchmarking data.

What is standardized mortality ratio (SMR) in APACHE II?

SMR = actual hospital deaths / APACHE II-predicted deaths. An SMR of 1.0 means actual mortality equals predicted mortality. SMR <1 indicates better-than-expected performance. SMR >1 warrants quality review. Most high-performing ICUs target SMR <0.8.

Can APACHE II be calculated if labs are unavailable?

Yes. If a specific lab test was not performed during the first 24 hours, assume a normal value (score = 0) for that parameter. Do not assume worst-case values for missing labs. Document that certain parameters were not measured.

What is the maximum APACHE II score?

The theoretical maximum is 71. In practice, scores above 40 are rare because achieving maximum points on all 12 physiologic variables simultaneously is uncommon. Scores above 35 are associated with very high mortality.

Online Medical Tools — APACHE II Score

Printed on 8/14/2026

For informational purposes only. This is not medical advice.

Emergency

APACHE II Score

The Acute Physiology and Chronic Health Evaluation II (APACHE II) is the most widely used ICU severity scoring system. It uses 12 acute physiological variables (worst values in first 24 hours of ICU admission), age points, and chronic health points to generate a score from 0 to 71. One critical component is the Glasgow Coma Scale, which assesses level of consciousness. Higher scores correlate with higher predicted mortality and help guide clinical decision-making, resource allocation, and benchmarking.

Formula: APACHE II = Acute Physiology Score + Age Points + Chronic Health Points (0–71)

Use worst values from the first 24 hours of ICU admission.

Temperature

°C

Mean Arterial Pressure

mmHg

Heart Rate

bpm

Respiratory Rate

/min

Oxygenation Score

Arterial pH

Sodium

mEq/L

Potassium

mEq/L

Creatinine

mg/dL

Hematocrit

White Blood Cell Count

×10³/mm³

Glasgow Coma Scale

Age

years

Chronic Health Points

Acute Renal Failure

Save your results with a free account

Keep a history of calculations, favorite tools, and access your dashboard anytime.

Create free account Sign in

How It Works

Record worst physiologic values in first 24h of ICU admission

Within the first 24 hours of ICU admission, record the most abnormal (worst) values for 12 physiologic variables: temperature, mean arterial pressure, heart rate, respiratory rate, oxygenation (A-a gradient or PaO₂), arterial pH, serum sodium, serum potassium, creatinine, hematocrit, white blood cell count, and Glasgow Coma Scale. Each variable is scored 0–4 based on deviation from normal. Use pre-sedation GCS if the patient was intubated after admission.

Add age points and chronic health points

Add age points (0–6): age <44 = 0, 45–54 = 2, 55–64 = 3, 65–74 = 5, ≥75 = 6. Add chronic health points (0 or 2 or 5) for severe pre-existing organ insufficiency: cirrhosis, portal hypertension, hepatic failure, lymphoma, metastatic cancer, leukemia, AIDS, or chronic immunosuppression. Elective post-op patients with these conditions receive 2 points; emergency post-op or non-operative patients receive 5 points.

Interpret the total and estimate predicted mortality

Sum all components (range 0–71). Higher scores predict higher ICU mortality. Approximate in-hospital mortality by score: 0–9 ≈ <10%; 10–14 ≈ 15%; 15–19 ≈ 25%; 20–24 ≈ 40%; 25–29 ≈ 55%; ≥30 ≈ 75%+. Use the score for ICU benchmarking (standardized mortality ratio = actual/expected deaths), clinical trial stratification, and prognostication discussions — not as a sole basis for individual treatment decisions.

Who Uses the APACHE II Score

ICU Admission Severity Documentation

Intensivists, critical care nurses

Calculate APACHE II at ICU admission to objectively document illness severity. The score establishes a baseline that supports goals-of-care communication with families, justifies ICU resource utilization, and meets regulatory documentation requirements. Pair with [SOFA Score](/tools/sofa-score) for ongoing daily organ failure tracking.

ICU Performance Benchmarking

ICU medical directors, quality officers

Calculate standardized mortality ratio (SMR = observed/expected deaths) using APACHE II-predicted mortality. An SMR <1 indicates better-than-expected performance; SMR >1 warrants quality review. National benchmarking databases (Project IMPACT, APACHE Outcomes) use APACHE II for inter-ICU comparisons. Track SMR trends quarterly to guide quality improvement initiatives.

Clinical Trial Stratification

Critical care researchers, clinical trial coordinators

APACHE II is the most commonly used inclusion and stratification criterion in critical care clinical trials. Required APACHE II thresholds ensure enrolled patients have comparable disease severity across study arms. Many landmark ICU trials (ARDSnet, ProCESS, ARISE) used APACHE II for eligibility screening and subgroup analysis.

Prognostication and Family Communication

Intensivists, palliative care teams

Use APACHE II-predicted mortality alongside clinical trajectory to frame prognostic conversations with families. Frame predictions probabilistically: 'Patients with scores like this typically have a 40–55% chance of surviving to hospital discharge, though individual outcomes vary.' Combine with [SOFA Score](/tools/sofa-score) trend for a more complete picture.

Mass Casualty and Resource Triage

Disaster medicine teams, hospital command

In mass casualty events where ICU beds are limited, APACHE II helps prioritize patients most likely to benefit from intensive care. Combined with reversibility assessment and patient wishes, it supports ethically defensible triage decisions when resources are constrained.

Sepsis and ARDS Research Applications

Critical care researchers, fellows

APACHE II is widely cited in sepsis, ARDS, and multi-organ failure literature for case-mix adjustment. Understanding APACHE II scoring is essential for interpreting published ICU outcomes data. Note: APACHE IV and SAPS III have improved calibration for modern ICU populations.

Pro Tips

APACHE II is designed for ICU admission only — not later time points

APACHE II is calculated once at ICU admission using first-24-hour worst values. It is a static snapshot, not a daily tracker. For ongoing organ failure monitoring, use [SOFA Score](/tools/sofa-score) serially. Recalculating APACHE II mid-ICU stay is not valid and will produce misleading results.

Use the worst (most abnormal) value — even if it was transient

The scoring rule is explicit: use the worst value recorded during the first 24 hours, regardless of how briefly it was abnormal. If temperature peaked at 39.8°C for 30 minutes, use 39.8°C. If pH dropped to 7.20 before correcting, use 7.20. This captures maximum physiologic derangement, which is what the model was trained on.

GCS in sedated patients: use pre-sedation GCS

If the patient was intubated and sedated before ICU admission, document the GCS assessed in the ED or pre-intubation. If this is unavailable and the patient is now sedated, many centers record GCS as 15 (assumed normal baseline) unless neurological dysfunction was present before sedation. Document your approach consistently.

Creatinine score doubles if acute renal failure is present

This is the most commonly missed scoring detail. If the patient has ACUTE renal failure (not chronic kidney disease), the creatinine score is doubled. A creatinine of 3.5 mg/dL scores 4 points normally but 8 points with ARF. Ensure you differentiate acute from chronic renal dysfunction when scoring.

Chronic health points: 2 vs 5 depends on surgery type

Patients with qualifying chronic conditions (cirrhosis, hepatic failure, portal hypertension, leukemia, lymphoma, metastatic cancer, AIDS, chronic immunosuppression) receive 2 points if they are elective post-operative patients, or 5 points if they are emergency post-operative or non-operative patients. This distinction significantly impacts the score.

APACHE II predicts hospital mortality — not ICU mortality alone

The original Knaus 1985 publication predicted hospital mortality (not 28-day or ICU mortality). Many publications now reference ICU mortality or 28-day mortality, which differ from hospital mortality. When benchmarking or comparing across studies, ensure the mortality definition is consistent.

SOFA is preferred for Sepsis-3 diagnosis and daily tracking

[SOFA Score](/tools/sofa-score) was chosen for Sepsis-3 (2016) because it can be calculated serially and tracks organ failure trajectory. APACHE II is a one-time admission snapshot. Both tools serve different clinical purposes — use SOFA for sepsis diagnosis and daily ICU monitoring, APACHE II for admission severity and benchmarking.

Modern calibration: actual mortality is often lower than APACHE II predicts

APACHE II was trained on 1980s ICU data. Improvements in care since then (lung-protective ventilation, early sepsis bundles, improved monitoring) mean many ICUs now have actual mortality significantly lower than APACHE II-predicted rates. SMR <1 is common in high-performing ICUs, reflecting improved care rather than easier case mix.

Missing labs: use normal values, not worst possible values

If a lab test was not obtained during the first 24 hours (e.g., ABG was never drawn), assume a normal value for that parameter. Do NOT assume the worst possible value. This is a common calculation error that inflates the score. The model's validity depends on this approach.

Temperature scoring: abnormal in both directions

APACHE II scores both fever and hypothermia. A temperature of 36°C (slightly low) and 39°C (slightly high) both score 1 point. The most abnormal direction scores highest. Hypothermia (<32°C) scores 4 points, the same as severe hyperthermia. Hypothermic patients are often more severely ill than their other parameters suggest.

Common Questions About Your Results

APACHE II is specifically designed for the first 24 hours of ICU admission. Calculating it at 48 hours using later values produces a different score that is not validated and cannot be compared to published mortality predictions. If the score was not calculated at admission, document this limitation. For ongoing severity assessment after 24 hours, use serial [SOFA Score](/tools/sofa-score) instead.

Use the GCS assessed BEFORE sedation and intubation if documented. Check the ED note, nursing notes, or paramedic documentation for pre-intubation neurological status. If no pre-sedation GCS is available, document the baseline as GCS 15 (assumed normal) unless there was evidence of neurological dysfunction prior to sedation. Some institutions have local policies for this — follow your ICU protocol.

APACHE II-predicted mortality is a population estimate, not an individual prediction. A 40% predicted mortality means that among 100 patients with this score and diagnosis, approximately 40 would be expected to die. For any individual patient, the actual probability of death is either 0% or 100% — they either survive or they don't. Use the score as a communication framework, not a deterministic prognosis.

Evidence-Based Methodology

APACHE II published by Knaus et al. (Crit Care Med 1985) from 5,815 ICU patients across 13 hospitals. C-statistic ~0.85 for in-hospital mortality prediction. APACHE IV (Zimmerman et al., Crit Care Med 2006) showed improved discrimination (C-statistic 0.88) using modern calibration data. SOFA score (Singer et al., JAMA 2016) incorporated into Sepsis-3 definitions for organ dysfunction. APACHE II remains the global standard for ICU benchmarking and clinical trial case-mix stratification.

Clinical Content Trust

Last reviewed:: April 21, 2026
Guideline version:: General evidence framework v2026.04
Source set version:: Primary-source set v1

How to Interpret Your Result

The APACHE II score ranges from 0 to 71, with higher scores indicating greater severity of illness and higher predicted hospital mortality. A score of 0–9 generally corresponds to a predicted mortality of less than 10%. Scores of 10–19 correspond to approximately 10–25% mortality. Scores of 20–29 suggest mortality in the range of 40–50%, and scores above 30 carry predicted mortality rates exceeding 50%, often reaching 70–80% or higher.

The score is most meaningful when interpreted in the context of the patient's specific diagnosis, as mortality predictions vary significantly by disease category. For example, the same APACHE II score may carry different prognostic implications in a post-surgical patient compared to a patient with septic shock. The score should be used alongside clinical judgment, not as a sole determinant of care decisions.

When to Use This Tool

The APACHE II score should be calculated within the first 24 hours of ICU admission using the worst (most abnormal) physiological values recorded during that period. It is primarily used for ICU quality benchmarking, comparing case-mix severity across units, and stratifying patients in clinical research trials.

Clinicians also use the APACHE II score to support goals-of-care discussions with patients and families, providing an objective framework for communicating illness severity. It can help guide resource allocation decisions and is sometimes used as an inclusion criterion in clinical trials to ensure comparable disease severity between treatment groups.

Limitations

APACHE II was developed and validated in 1985 using data from 5,815 ICU admissions. Advances in critical care since then — including early goal-directed therapy, lung-protective ventilation, and modern sepsis management — mean that actual mortality rates for a given APACHE II score are often lower than the original predictions. More recent scoring systems (APACHE III, IV, [SOFA](/tools/sofa-score), SAPS) may provide better calibration for current ICU populations.

The score uses the worst values in the first 24 hours, which means it cannot be calculated at the time of ICU admission and does not capture clinical trajectory or response to treatment. Patients who deteriorate after the first 24 hours will not have this reflected in their APACHE II score. Additionally, the chronic health component relies on subjective assessment of pre-existing organ insufficiency, introducing some variability between assessors.

The score was not designed to guide individual treatment decisions (such as whether to withdraw care) and should never be used as the sole basis for such decisions. It performs best as a tool for population-level risk stratification and quality assessment.

Disclaimer: This tool is for educational and informational purposes only. It is not a substitute for professional medical advice, diagnosis, or treatment. Always consult a qualified healthcare provider with questions about your health.

Changelog

April 21, 2026 · trust-baseline
Clinical trust metadata enabled for this tool page with structured review/version fields.

Related Tools