The impact of illness severity evolution measured with LOD, SOFA, APACHE II and SAPS II scores on the development of ICU acquired infections

Abstract
Several studies were intended to evaluate ICU risk factors, but only few integrated the impact of illness severity variation in their study.
The aim of the present study is to explore the illness severity variation for patients having contracted an ICU-acquired infection and to check if this variation correlates to the nosocomial episode. The study is mono-centric, retrospective and non interventional. It is a case control type with matching approach 1 case vs. 1 control. It is based on a total of 250 patients, at least of 16-year-old, who spent a minimum of 72 hours in the ICU of Timone University Hospital.
The severity of the illness was measured according to the following scores: Logistic Organ Dysfunction (LOD), Simplified Acute Physiology Score (SAPS), Sequential Organ Failure Assessment (SOFA) and Acute Physiology and Chronic Health Evaluation (APACHE).
These scores were calculated retrospectively each day, from the day of admission in ICU to the day of discharge or death, except for APACHE score, which is interesting only for the first 24h of admission. The ability of the models for predicting ICU mortality was determined by examining their discrimination. Forty-six patients developed one or more nosocomial episodes, and 65 died. The matching of cases was successful for 71.73% of them. All SOFA scores values are higher in the case’s group: IS (6.652±2.946 cases vs. 5.795±2.494 controls), H48 (6.652±2.767 cases vs. 5.077±2.082 controls), H72 (6.957±2.944 cases vs. 4.538±2.258 controls).
It is quite similar for the LOD score IS (6.087±2.889 cases vs. 5.821±2.72 controls), H48 (5.870±2.864 cases vs. 5.154±1.94 controls), H72 (5.761±2.677 cases vs. 5.256±2.074 controls). As far as APACHE II score is concerned, it counts: 20.478±7.938 cases vs. 24.23±7.938 controls. Among the four documented scores, only SOFA H48 has foreseen a nosocomial risk.