Intensive Care Infection Score (ICIS) is an Early Marker for Infection in Time of Admission to Intensive Care Units.

Abstract

Background: Despite all the effort, infections remain one of the major causes of morbidity and mortality in clinically ill patients, and novel diagnostic markers detecting infections in early stages are searched for. Intensive Care Infection Score (ICIS) was developed as such a marker.

Methods: A total of 102 patients admitted to intensive care units (ICU) in the University Hospital Hradec Kralove, Czech Republic were enrolled in this study. ICIS along with relevant biochemical markers (procalcitonin, C-reactive protein and Interleukin 6) was analyzed on the day of the admission. Individual parameters used to calculate ICIS were analyzed too. Infection was subsequently confirmed in 30 patients.

Results: ICIS predicted infections with the highest AUC (0.958) of all analyzed markers. The cut-off value (< 4) was selected as the value with the highest Youden index, and it predicted sepsis with high specificity (84.2%) and sensitivity (93.3%). Negative predictive value was very high too (96.8%). Positive predictive value was 71.8%.

Conclusions: ICIS is a reliable, cheap, fast, and simply interpretable score for the early identification of infection in patients admitted to ICUs. ICIS ≥ 4 predicts infection with high sensitivity, specificity, and negative predictive value.