APACHE II Score Calculator

Q: Which oxygenation value does APACHE II use?

It depends on the FiO₂. At an FiO₂ of 0.5 or above, the alveolar–arterial oxygen gradient is scored (≥500 → 4, 350–499 → 3, 200–349 → 2, below 200 → 0). Below an FiO₂ of 0.5, the PaO₂ itself is scored (above 70 → 0, 61–70 → 1, 55–60 → 3, below 55 → 4). Never score both.

Q: Why are creatinine points sometimes doubled?

Knaus specifies "double point score for acute renal failure". A creatinine of 2.1 mg/dL scores 3 points in chronic renal impairment and 6 points if the failure is acute, which is a two-band swing in the total on its own.

Q: How does the Glasgow Coma Scale contribute?

The contribution is 15 minus the actual GCS, so a normal GCS of 15 adds nothing and a GCS of 3 adds 12 points — more than any other single variable can. Use the score from before sedation, even if that reading falls outside the 24-hour scoring window.

Q: What is the maximum APACHE II score?

71. Ten variables can contribute 4 each, creatinine can contribute 8 after doubling, the GCS up to 12, age up to 6 and chronic health up to 5.

Q: Is APACHE II still accurate?

Its ability to rank patients by severity has held up; its calibration has not. The APACHE IV authors recommended it "no longer be used to compare observed and predicted mortality", and it is widely reported to overestimate mortality against modern outcomes — though at least one contemporary cohort found it under-predicting instead. Treat the percentage as a severity index, not a prognosis.

Score all twelve APACHE II physiological variables plus age and chronic health, with the oxygenation branch and the acute-renal-failure doubling handled for you.

Last updated 17 Sep 2026

APACHE II Score

12 variables + age + chronic health → 0–71

Temperature — rectal (°C)

Mean arterial pressure (mmHg)

Heart rate — ventricular response (per minute)

Respiratory rate — ventilated or not (per minute)

FiO₂ as a fractionAt 0.5 or above the score uses the alveolar–arterial gradient; below 0.5 it uses the PaO₂.

PaO₂ (mmHg)1 kPa = 7.5 mmHg.

PaCO₂ (mmHg)Only used to compute the alveolar–arterial gradient at FiO₂ 0.5 or above.

Arterial pHIf no arterial gas is available, Knaus allows venous bicarbonate instead — see the table below.

Serum sodium (mmol/L)

Serum potassium (mmol/L)

Serum creatinine (mg/dL)88.4 µmol/L is 1.0 mg/dL. 2.0 mg/dL is about 177 µmol/L.

Acute renal failureKnaus: “Double point score for acute renal failure.”

Haematocrit (%)

White cell count (×10⁹/L)×10⁹/L is the same number as thousands per mm³, the unit Knaus printed.

Glasgow Coma Scale — before sedationContributes 15 − GCS. Use the pre-sedation score even if it falls outside the 24-hour window.

Age (years)

Chronic health points

Admission diagnostic category — for the mortality equation onlyKnaus’s diagnostic category weight. It does not change the points total.

Emergency post-operative surgery — for the mortality equation only

26pointsExample

68-year-old, septic shock from pneumonia. Temperature 39.5 °C, MAP 58 mmHg, heart rate 124, respiratory rate 32, FiO₂ 0.6 with PaO₂ 70 and PaCO₂ 38 mmHg, pH 7.28, sodium 133, potassium 4.6, creatinine 2.1 mg/dL with acute renal failure, haematocrit 33%, white cells 18.4 ×10⁹/L, GCS 13, no chronic organ insufficiency

Scoring

APACHE II = Acute Physiology Score (12 variables, 0–4 each) + age points (0–6) + chronic health points (0, 2 or 5), maximum 71

Oxygenation: at FiO₂ ≥0.5 the A–a gradient, (FiO₂ × 713) − (PaCO₂ ÷ 0.8) − PaO₂; below 0.5 the PaO₂ itself
Creatinine: points doubled if the renal failure is acute
GCS: contributes 15 − GCS, so a GCS of 3 contributes 12
Age points: ≤44 → 0, 45–54 → 2, 55–64 → 3, 65–74 → 5, ≥75 → 6
Chronic health: 5 if non-operative or emergency post-operative, 2 if elective post-operative, and only with severe chronic organ insufficiency or immunocompromise
Predicted mortality: ln(R ÷ (1 − R)) = −3.517 + (0.146 × score) + 0.603 if emergency post-operative + diagnostic category weight

Worked example

68-year-old, septic shock from pneumonia. Temperature 39.5 °C, MAP 58 mmHg, heart rate 124, respiratory rate 32, FiO₂ 0.6 with PaO₂ 70 and PaCO₂ 38 mmHg, pH 7.28, sodium 133, potassium 4.6, creatinine 2.1 mg/dL with acute renal failure, haematocrit 33%, white cells 18.4 ×10⁹/L, GCS 13, no chronic organ insufficiency
Temperature 39.5 °C → 39–40.9 → 3; MAP 58 → 50–69 → 2; heart rate 124 → 110–139 → 2; respiratory rate 32 → 25–34 → 1
FiO₂ is 0.6, so the gradient ladder applies: A–a = (0.6 × 713) − (38 ÷ 0.8) − 70 = 427.8 − 47.5 − 70 = 310.3 mmHg → 200–349 → 2
pH 7.28 → 7.25–7.32 → 2; sodium 133 → 130–149 → 0; potassium 4.6 → 3.5–5.4 → 0
Creatinine 2.1 mg/dL → 2.0–3.4 → 3, doubled for acute renal failure → 6
Haematocrit 33% → 30–45.9 → 0; white cells 18.4 → 15–19.9 → 1; GCS 13 → 15 − 13 = 2
Acute Physiology Score = 3+2+2+1+2+2+0+0+6+0+1+2 = 21
Age 68 → 65–74 → 5; chronic health 0 → APACHE II = 21 + 5 + 0 = 26 points
Mortality equation, sepsis category (weight 0.113): −3.517 + (0.146 × 26) + 0.113 = 0.392, so predicted mortality = e^0.392 ÷ (1 + e^0.392) = 59.7%

The Acute Physiology Score — all twelve variables

Adapted from Knaus WA et al, Crit Care Med 1985;13(10):818–829. Bicarbonate is a substitute for arterial pH when no blood gas is available, not a thirteenth variable — this calculator scores the pH, so use the bicarbonate row by hand if that is all you have.
Variable	+4	+3	+2	+1	0	+1	+2	+3	+4
Temperature, rectal (°C)	≥41	39–40.9	—	38.5–38.9	36–38.4	34–35.9	32–33.9	30–31.9	≤29.9
Mean arterial pressure (mmHg)	≥160	130–159	110–129	—	70–109	—	50–69	—	≤49
Heart rate	≥180	140–179	110–139	—	70–109	—	55–69	40–54	≤39
Respiratory rate	≥50	35–49	—	25–34	12–24	10–11	6–9	—	≤5
Oxygenation, FiO₂ ≥0.5: A–a gradient	≥500	350–499	200–349	—	<200	—	—	—	—
Oxygenation, FiO₂ <0.5: PaO₂ (mmHg)	—	—	—	—	>70	61–70	—	55–60	<55
Arterial pH	≥7.70	7.60–7.69	—	7.50–7.59	7.33–7.49	—	7.25–7.32	7.15–7.24	<7.15
Serum sodium (mmol/L)	≥180	160–179	155–159	150–154	130–149	—	120–129	111–119	≤110
Serum potassium (mmol/L)	≥7.0	6.0–6.9	—	5.5–5.9	3.5–5.4	3.0–3.4	2.5–2.9	—	<2.5
Serum creatinine (mg/dL) — double in acute renal failure	≥3.5	2.0–3.4	1.5–1.9	—	0.6–1.4	—	<0.6	—	—
Haematocrit (%)	≥60	—	50–59.9	46–49.9	30–45.9	—	20–29.9	—	<20
White cell count (×10⁹/L)	≥40	—	20–39.9	15–19.9	3–14.9	—	1–2.9	—	<1
Glasgow Coma Scale	Contributes 15 − GCS (0 to 12 points)
Serum HCO₃⁻ (mmol/L) — only if no arterial gas	≥52	41–51.9	—	32–40.9	22–31.9	—	18–21.9	15–17.9	<15

Age points and chronic health points

Chronic organ insufficiency means biopsy-proven cirrhosis with documented portal hypertension, NYHA class IV heart failure, severe chronic respiratory disease with documented hypoxia, hypercapnia or pulmonary hypertension, or chronic dialysis. Immunocompromise includes immunosuppressive therapy and long-term high-dose corticosteroids.
Component	Criterion	Points
Age	≤44 years	0
Age	45–54	2
Age	55–64	3
Age	65–74	5
Age	≥75	6
Chronic health	No severe chronic organ insufficiency and not immunocompromised	0
Chronic health	Elective post-operative admission, with chronic insufficiency or immunocompromise	2
Chronic health	Non-operative or emergency post-operative admission, with chronic insufficiency or immunocompromise	5

The oxygenation branch, and why the percentage is the 1985 model’s opinion

APACHE II is deliberately large. Twelve physiological variables scored 0 to 4, age points and chronic health points add to a maximum of 71, and the size is the design: no single derangement dominates, and a patient has to be sick in several systems at once to reach the top bands. The Acute Physiology Score is shown separately beside the total because it is the part that can change in twenty-four hours — age and chronic health cannot.

One variable is got wrong far more often than the rest, and it is oxygenation. APACHE II does not score a single oxygen number. At an FiO₂ of 0.5 or above it scores the alveolar–arterial gradient, on a ladder running ≥500, 350–499, 200–349 and below 200. Below an FiO₂ of 0.5 it scores the PaO₂ itself, on a completely different ladder running above 70, 61–70, 55–60 and below 55. The two ladders have different lengths and run in opposite directions, so scoring a patient in 60% oxygen off the PaO₂ row understates the score, by as much as four points. Knaus’s own instruction is blunt: do not score for both. This calculator takes the FiO₂, PaO₂ and PaCO₂, computes the gradient with the same equation as the A–a gradient page and picks the ladder from the FiO₂, so the branch cannot be taken wrongly. Two smaller rules catch people out as well: creatinine points double when the renal failure is acute, and the neurological contribution is 15 − GCS, so a patient with a GCS of 3 contributes 12 points before anything else is counted.

The predicted mortality shown beside the total comes from Knaus’s logistic equation — the logit is −3.517 plus 0.146 per point, plus 0.603 for emergency post-operative admission, plus a weight for the admission diagnosis. Those weights range from −3.353 for a drug overdose to +0.891 for respiratory failure caused by a neoplasm, so the same 20-point score can predict wildly different outcomes depending on why the patient is there. That is a feature of the model, and it is also the reason a bare APACHE II number quoted without its diagnostic category tells you less than it appears to.

Read the percentage as a 1985 opinion. The model was derived from 5,030 admissions to thirteen American intensive care units between 1979 and 1982, and its calibration has not aged well. The authors of APACHE IV wrote that “the accuracy of older prognostic models such as APACHE II has deteriorated over time” and recommended that it “no longer be used to compare observed and predicted mortality”. Moreno and Nassar go further, saying APACHE II “generally overestimates mortality in many scenarios in which it is applied” — on 1985 expectations, “almost any ICU today would be considered ‘high performance'”. The evidence is not unanimous, and the page would be dishonest if it pretended otherwise: a 2016 Polish mixed-ICU cohort found the opposite, an observed-to-predicted ratio of 1.12, with observed mortality of 35.6% against a median APACHE II prediction of 25.8%. What everyone agrees on is that the discrimination has held up better than the calibration — APACHE II still ranks patients by severity usefully, and it is that ranking, rather than the percentage, that the score is worth using for. This supports a clinician’s judgement rather than replacing it. It is arithmetic on the figures entered, and it knows nothing about the patient in front of you.

Frequently asked questions

Which oxygenation value does APACHE II use?

It depends on the FiO₂. At an FiO₂ of 0.5 or above, the alveolar–arterial oxygen gradient is scored (≥500 → 4, 350–499 → 3, 200–349 → 2, below 200 → 0). Below an FiO₂ of 0.5, the PaO₂ itself is scored (above 70 → 0, 61–70 → 1, 55–60 → 3, below 55 → 4). Never score both.

Why are creatinine points sometimes doubled?

Knaus specifies “double point score for acute renal failure”. A creatinine of 2.1 mg/dL scores 3 points in chronic renal impairment and 6 points if the failure is acute, which is a two-band swing in the total on its own.

How does the Glasgow Coma Scale contribute?

The contribution is 15 minus the actual GCS, so a normal GCS of 15 adds nothing and a GCS of 3 adds 12 points — more than any other single variable can. Use the score from before sedation, even if that reading falls outside the 24-hour scoring window.

What is the maximum APACHE II score?

71. Ten variables can contribute 4 each, creatinine can contribute 8 after doubling, the GCS up to 12, age up to 6 and chronic health up to 5.

Is APACHE II still accurate?

Its ability to rank patients by severity has held up; its calibration has not. The APACHE IV authors recommended it “no longer be used to compare observed and predicted mortality”, and it is widely reported to overestimate mortality against modern outcomes — though at least one contemporary cohort found it under-predicting instead. Treat the percentage as a severity index, not a prognosis.

Related calculators

References

Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. 1985;13(10):818–29.
Zimmerman JE, Kramer AA, McNair DS, Malila FM. Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today’s critically ill patients. Crit Care Med. 2006;34(5):1297–310.
Moreno RP, Nassar AP Jr. Is APACHE II a useful tool for clinical research? Rev Bras Ter Intensiva. 2017;29(3):264–7.
Rapsang AG, Shyam DC. Scoring systems in the intensive care unit: a compendium. Indian J Crit Care Med. 2014;18:220–8.
Czajka S, Ziębińska K, Marczenko K, et al. Validation of APACHE II, APACHE III and SAPS II scores in in-hospital and one year mortality prediction in a mixed intensive care unit in Poland. BMC Anesthesiol. 2020;20:296.

Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.