On the Use of Graphical Models to Study ICU Outcome Prediction in Septic Patients Treated with Statins

Sepsis is a common pathology in Intensive Care Units (ICU). At its most acute phase, namely septic shock, mortality rates can be as high as 60%. Early administration of antibiotics is known to be crucial for ICU outcomes. In particular, statins, a class of drug usually associated to the regulation of the biosynthesis of cholesterol, have been shown to present good anti-inflammatory properties. In this brief study, we hypothesized that the pre-admission use of statins should improve ICU outcomes, reducing the mortality rate. We tested this hypothesis in a prospective study with patients admitted with severe sepsis and multi-organ failure at the ICU of Vall d’Hebron University Hospital (Barcelona, Spain). Outcome was predicted using statistic algebraic models and the Bayesian Networks that can naturally be derived from these models.