Validation of a novel molecular host response assay to diagnose infection in hospitalized patients admitted to the ICU with acute respiratory failure

Abstract

Objectives: Discrimination between infectious and noninfectious causes of acute respiratory failure is difficult in patients admitted to the ICU after a period of hospitalization. Using a novel biomarker test (SeptiCyte LAB), we aimed to distinguish between infection and inflammation in this population. Design: Nested cohort study. Setting: Two tertiary mixed ICUs in the Netherlands. Patients: Hospitalized patients with acute respiratory failure requiring mechanical ventilation upon ICU admission from 2011 to 2013. Patients having an established infection diagnosis or an evidently noninfectious reason for intubation were excluded. Interventions: None. Measurement and Main Results: Blood samples were collected upon ICU admission. Test results were categorized into four probability bands (higher bands indicating higher infection probability) and compared with the infection plausibility as rated by post hoc assessment using strict definitions. Of 467 included patients, 373 (80%) were treated for a suspected infection at admission. Infection plausibility was classified as ruled out, undetermined, or confirmed in 135 (29%), 135 (29%), and 197 (42%) patients, respectively. Test results correlated with infection plausibility (Spearman's rho 0.332; p < 0.001). After exclusion of undetermined cases, positive predictive values were 29%, 54%, and 76% for probability bands 2, 3, and 4, respectively, whereas the negative predictive value for band 1 was 76%. Diagnostic discrimination of SeptiCyte LAB and C-reactive protein was similar (p = 0.919). Conclusions: Among hospitalized patients admitted to the ICU with clinical uncertainty regarding the etiology of acute respiratory failure, the diagnostic value of SeptiCyte LAB was limited.