The utility of procalcitonin for identifying secondary infections in patients with influenza or COVID-19 receiving extracorporeal membrane oxygenation

Abstract

Background: Identifying secondary infections in patients receiving extracorporeal membrane oxygenation (ECMO) presents challenges due to the ECMO circuit’s influence on traditional signs of infection. Objectives: This study evaluates procalcitonin as a diagnostic marker for secondary infections in patients receiving ECMO with influenza or COVID-19 infection. Design: Single-center retrospective cohort study. Methods: All adult patients receiving veno-venous ECMO with underlying influenza or COVID-19 from November 2017 to October 2021 were included. Patient demographics, time receiving ECMO, culture data, and procalcitonin levels were examined. The first procalcitonin within 3 days of infection was compared to negative workups that were collected at least 10 days from the last positive culture. Furthermore, we compared procalcitonin levels by the type of pathogen and site of infection. Results: In this study, 84 patients with influenza or COVID-19 who received ECMO were included. A total of 276 procalcitonin labs were ordered in this cohort, with 33/92 (36%) of the secondary infections having an associated procalcitonin value. When comparing procalcitonin levels, there was no significant difference between the infection and negative workup groups [1 ng/mL (interquartile ranges, IQR: 0.4–1.2) versus 1.3 (0.5–4.3), p = 0.19]. Using 0.5 ng/mL as the cut-off, the sensitivity of procalcitonin was 67% and the specificity was 30%. In our cohort, the positive predictive value of procalcitonin was 14.5% and the negative predictive value was 84%. There was no difference in procalcitonin by type of organism or site of infection. Procalcitonin levels did not routinely decline even after an infection was identified. Conclusion: While procalcitonin is a proposed potential diagnostic marker for secondary infections in patients receiving ECMO, this single-center study demonstrated low sensitivity and specificity of procalcitonin in identifying secondary infections. Furthermore, there was no association of procalcitonin levels with etiology of infection when one was present. Procalcitonin should be used cautiously in identifying infections in veno-venous ECMO.