To develop a customized version of SAPS II for ICU patients with early severe sepsis, we selected 1130 patients from the European-North American Study database who had been in intensive care for at least 24 h (LeG.al! eL§L 1996). These patients had severe sepsis as defined by the criteria of the American College of Chest
Physicians and the Society of Critical Care Medicine (systemic inflammatory response syndrome in response to infection plus hypotension, hypoperfusion, or multiple organ dysfunction). In patients with severe sepsis, mortality was higher (48.0 versus 19 per cent among other patients) and 28-day survival was lower. The customized SAPS II was well calibrated (p = 0.92 for the goodness-of-fit test) and discriminated well (area under the ROC curve, 0.78). Performance in the validation sample was equally good (p = 0.85 for the goodness-of-fit test; area under the ROC curve, 0.79).
The formula for the probability of death for a patient with early severe sepsis is cfe + p](SAPS 11 score) 1 + cpD + pi (SAPS n score)
where b0 = -3.5524 and b1 = 0.0694. For instance with a SAPS II score of 40 the probability of hospital death is 0.315 (the standard formula would have given a probability of 0.247).
Customization provides a simple technique for applying existing models to a subgroup of patients. Accurate assessment of the probability of hospital mortality is a useful adjunct to clinical trials.
Was this article helpful?