Abstract / Summary
Identification of infection in critically ill patients remains challenging, as conventional biomarkers such as C-reactive protein (CRP), procalcitonin (PCT) and white blood cell count (WBC) frequently lack diagnostic accuracy. The Intensive Care Infection Score (ICIS), a host-response biomarker derived from hematological parameters, reflects both innate and adaptive immune response. To compare diagnostic performance of ICIS, CRP, PCT, and WBC, expert adjudication was used to determine presence or absence of infection. Diagnostic accuracy parameters including receiver operating characteristic curve analysis were calculated. 770 samples from 382 critically ill patients were included. ICIS demonstrated the highest diagnostic accuracy with an area under the curve (AUC) of 0.914 (95% CI 0.892–0.933), compared with CRP (AUC 0.889; 95% CI 0.865–0.910), PCT (AUC 0.847; 95% CI 0.819–0.871), and WBC (AUC 0.697; 95% CI 0.663–0.729). At the optimal cut-off value of > 3, ICIS achieved a sensitivity of 80.7% and specificity of 89.3%. ICIS showed the lowest false-positive rate (10.7%) and a negative predictive value of 88.7%. ICIS demonstrated good diagnostic performance in critically ill patients and can contribute to antimicrobial stewardship as a novel host-response diagnostic.