ABSTRACT
Low procalcitonin (PCT) concentrations (<0.5ng/mL) can facilitate exclusion of bacterial co-infection in viral infections, including COVID-19. However, costs associated with PCT measurement preclude universal adoption, indicating a need to identify settings where PCT provides clinical information beyond that offered by other inflammatory markers, such as C-reactive protein (CRP) and white cell count (WCC). In an unselected cohort of 299 COVID-19 patients, we tested the hypothesis that PCT<0.5ng/mL was associated with lower levels of CRP and WCC. We demonstrated that CRP values below the geometric mean of the entire patient population had a negative predictive value for PCT<0.5ng/mL of 97.6% and 100% at baseline and 48 hours into admission respectively, and that this relationship was not confounded by intensive care admission or microbiological findings. CRP-guided PCT testing algorithms can reduce costs and support antimicrobial stewardship strategies in COVID-19.