🏥 APACHE II Calculator

APACHE II is the classic ICU severity score: twelve physiologic measurements from the first 24 hours after admission, plus points for age and chronic health, combined into a single number from 0 to 71. Higher scores track closely with hospital mortality, which made it the standard for comparing ICU populations and stratifying clinical trials.

This calculator implements the standard point tables: each variable scores 0 to 4 by its worst value, age adds 0 to 6, and chronic health adds 0, 2, or 5. The total maps to the published mortality bands. It is an educational implementation for learning the system, not a bedside prognosticator.

How to use this calculator

Enter the worst value of each variable during the first 24 hours of ICU admission. "Worst" means furthest from normal in either direction.

How the points work

Every variable is scored 0 to 4 against fixed bands, with both extremes penalized: a temperature of 41 and a temperature of 30 both score 4. Oxygenation uses PaO2 when FiO2 is under 50 percent and the A-a gradient above it. GCS contributes 15 minus the score directly.

Chronic health points apply only with documented organ insufficiency or immunocompromise: 5 for nonoperative or emergency postoperative patients, 2 for elective postoperative ones. Creatinine points double in acute renal failure.

Worked example

A 68-year-old admitted with sepsis, no chronic comorbidity. Worst first-24h values: temp 38.6 (1 point), MAP 72 (0), heart rate 118 (2), respiratory rate 28 (1), FiO2 40% with PaO2 68 (1), pH 7.31 (2), sodium 138 (0), potassium 4.8 (0), creatinine 1.6 (2), hematocrit 33 (0), white cells 16 (1), GCS 13 (2 points). Variable sum: 12. Age 68 adds 5. Chronic health adds 0. Total: 17.

A score of 17 falls in the 15 to 19 band, roughly 25 percent predicted hospital mortality on the nonoperative reference curve. That number describes populations, not this patient, but it quantifies what the bedside team already senses: a seriously ill patient.

What the score is good for

APACHE II shines at comparing groups: benchmarking ICUs, risk-adjusting trial arms, and studying whether a new therapy helps sicker or less-sick patients. It was never meant to predict one person's outcome, and using it that way is the classic misuse. Two patients with a score of 20 can have very different trajectories based on diagnosis, reversibility, and response to treatment.

Newer systems (APACHE III/IV, SAPS, SOFA for daily tracking) improved on it, but APACHE II endures in teaching because its tables are transparent: you can see exactly where every point came from. For acid-base variables in the score, our ABG interpreter explains the physiology, and the Aldrete score covers the other end of the acuity spectrum: recovery after anesthesia.

Common scoring pitfalls

The most frequent errors: using admission values instead of the worst 24-hour values, forgetting to double creatinine points in acute renal failure, scoring chronic health points without qualifying comorbidity, and using venous instead of arterial pH. Sedation also depresses GCS: score the best neurological function, not the sedated one, or note the limitation.

Missing variables default to normal (0 points), which systematically underestimates severity when data are sparse. A score computed from half a chart is a floor, not a measurement.

SOFA vs APACHE: snapshot vs movie

APACHE II photographs severity at admission; SOFA (Sequential Organ Failure Assessment) films it day by day. SOFA scores six organ systems from 0 to 4 daily, and the trend tells the story: falling SOFA means recovery, rising SOFA means trouble, regardless of where it started.

Clinicians use both without confusion: APACHE II for "how sick is this population," SOFA for "is this patient getting better." A patient admitted with APACHE II 25 whose SOFA drops from 12 to 5 over a week is winning; one admitted at 15 whose SOFA climbs from 6 to 14 is losing. The movie matters more than the snapshot.

APACHE in research: risk adjustment

Clinical trials in critical care live or die by risk adjustment, and APACHE II was built for it. Randomizing 500 septic patients means nothing if one arm is sicker; stratifying by APACHE II ensures the comparison is fair. ICU benchmarking works the same way: a unit with higher mortality might simply admit sicker patients, and APACHE-adjusted ratios reveal the truth.

The score's transparency is its research virtue. Unlike black-box models, anyone can recompute an APACHE II from the chart and audit the result. That auditability is why it survived decades of fancier competitors in the literature, even as bedside practice moved on.

For informational purposes only, not medical advice. Consult a healthcare professional.