SAPS II calculator

The Simplified Acute Physiology Score II, published by Le Gall, Lemeshow, and Saulnier in 1993, converts age, admission type, chronic disease, and twelve admission physiological measurements into an integer ICU severity score and a predicted hospital mortality probability. This calculator applies the original JAMA point tables and the logistic equation logit equals negative 7.7631 plus 0.0737 times the score plus 0.9971 times the natural log of score plus one, then maps that logit to a probability with the standard expit.

What SAPS II measures and how the Le Gall 1993 tables add up

The Simplified Acute Physiology Score II is an additive ICU severity system derived by Le Gall, Lemeshow, and Saulnier from 13,152 consecutive adult admissions across 137 medical and surgical intensive care units in twelve countries, and published in JAMA in 1993. Unlike APACHE II, SAPS II does not require a primary diagnosis to estimate hospital death: twelve physiological measurements recorded as the worst values during the first twenty-four hours, plus age, type of admission, and one chronic-disease category, are converted to integer points from published tables and summed.

Age contributes zero under forty years, seven from forty through fifty-nine, twelve from sixty through sixty-nine, fifteen from seventy through seventy-four, sixteen from seventy-five through seventy-nine, and eighteen at eighty or older. Heart rate, systolic blood pressure, temperature at or above thirty-nine degrees Celsius, the PaO2/FiO2 ratio when the patient is ventilated or on CPAP, twenty-four-hour urine output, blood urea nitrogen, white-cell count, potassium, sodium, bicarbonate, bilirubin, and the Glasgow Coma Scale each add a published weight that is larger when the value is more abnormal.

Admission type adds zero for scheduled surgery, six for a medical admission, and eight for unscheduled surgery. Chronic disease adds nine for metastatic cancer, ten for hematologic malignancy, and seventeen for AIDS, taking the single most severe category when more than one applies. The integer total is the SAPS II score.

Hospital mortality then follows the original logistic: logit equals negative 7.7631 plus 0.0737 times the score plus 0.9971 times the natural logarithm of the score plus one, and probability equals e to the logit over one plus e to the logit.

Input contract, units, validation, and the shape of a successful response

Age is years at last birthday. Heart rate is beats per minute. Systolic blood pressure is millimetres of mercury.

Temperature is degrees Celsius. Glasgow Coma Scale is the numeric total, typically three through fifteen; if the patient is sedated, use the score recorded before sedation, matching the original instructions. Urine output is litres in twenty-four hours, not millilitres: half a litre is 0.5, not 500.

Blood urea nitrogen is milligrams per decilitre; urea in millimoles per litre times 2.8 is the usual conversion to BUN. White-cell count is thousands per cubic millimetre, equivalent to 10^9 per litre. Potassium, sodium, and bicarbonate are milliequivalents per litre.

Bilirubin is milligrams per decilitre. Ventilated defaults to false and contributes zero oxygenation points when false. Unknown admission or chronic-disease tokens also fail.

How to read predicted mortality, where the calculator belongs, and what it does not decide

A SAPS II of about twenty-nine corresponds to roughly ten percent predicted hospital mortality in the original logistic, forty to about twenty-five percent, fifty-two to about fifty percent, sixty-four to about seventy-five percent, and seventy-seven to about ninety percent. Those figures are cohort probabilities from the 1993 European and North American sample, not a personal prognosis, and they do not replace serial examination, diagnosis, source control, or a clinician's view of comorbidities this table never captures.

Cardiac-surgery, burn, and paediatric patients were excluded from the derivation sample, so applying the logistic there is an extrapolation. Use the calculator for education, exam keys, research coding of adult ICU rows, and product prototypes that need a deterministic SAPS II step before handing off to licensed systems. Neighbouring tools such as the Pneumonia Severity Index, the Revised Cardiac Risk Index, or the Revised Trauma Score answer different search intents and should keep their own landings rather than being overloaded here.

Because every physiological input is checked for finiteness, a spreadsheet that leaks NaN or a blank cell fails closed instead of silently scoring a missing lab as normal.