Comparison between procalcitonin and C-reactive protein to predict blood culture results in ICU patients

Dear Editor, Biomarkers represent an essential tool for identification of patients developing infection and to determine their clinical severity. Procalcitonin (PCT) levels appeared to be correlated with the development of severe bacterial infections [1]. Thus, PCT systematic use has been proposed as part of the diagnostic tools and for monitoring treatment duration [2, 3], but not all of the potential benefits and limitations of PCT have been investigated. We retrospectively performed a case-control study analyzing all patients with positive blood cultures (BCs) in the period of January 2017 to December 2017 at a 1100-bed teaching hospital in Italy and investigating the correlation between PCT and C-reactive protein (CRP) values (± 24 h from BC collection) in pathogens causing bloodstream infections. The study flowchart is presented in Additional file 1: Figure S1. During the study period, 1296 positive BCs were retrieved; of these, 258 (19.9%) episodes were recorded in the intensive care unit (ICU) and were included in the study. Moreover, 213 patients hospitalized in ICU with negative BC were used as control. Finally, 471 ICU patients were analyzed. Clinical characteristic and outcome of patients, according to BC results, are reported in Additional file 1: Table S1. As reported in Fig. 1, PCT concentrations (in nanograms per milliliter) were 25.1 ± 19.9 in patients with Gram-negative (GN) etiology, 29.9 ± 13.2 for Enterobacteriaceae, 8.9 ± 7.5 for Gram-positive (GP), and 2.1 ± 1.8 for fungi. Finally, in Additional file 1: Figure S1, receiver operating characteristic curves showed an area under the curve of 0.7 (95% confidence interval (CI) 0.62–0.77, P < 0.001) for PCT and 0.45 (95% CI 0.37–0.54, P = 0.32) for CRP among GN isolates, 0.74 (95% CI 0.67–0.81, P <0.001) for PCT and 0.49 (95% CI 0.4–0.57, P = 0.82) for CRP among Enterobacteriaceae, 0.46 (95% CI 0.39–0.53, P = 0.38) for PCT and 0.41 (95% CI 0.33–0.48, P = 0.01) for CRP among GP isolates, and 0.64 (95% CI 0.46–0.83, P = 0.22) for PCT and 0.59 (95% CI 0.45–0.73, P = 0.43) for CRP among fungi. Finally, logistic regression analysis showed that PCT values of more than 0.5 ng/mL and more than 10 ng/mL were independently associated with BCs positive for Enterobacteriaceae. Our data confirmed the previous observations about the role of PCT and CRP in predicting BC results in critically ill patients [4, 5]. Of interest, CRP was not able to predict BC results, whereas PCT values correlated with GN bacteremia and, among them, specifically identified Enterobacteriaceae. High PCT values (> 10 ng/mL) were independently associated with Enterobacteriaceae isolation. Even with the limitation of a single-center experience, these results might be useful to determine another role


Dear Editor,
Biomarkers represent an essential tool for identification of patients developing infection and to determine their clinical severity. Procalcitonin (PCT) levels appeared to be correlated with the development of severe bacterial infections [1]. Thus, PCT systematic use has been proposed as part of the diagnostic tools and for monitoring treatment duration [2,3], but not all of the potential benefits and limitations of PCT have been investigated.
We retrospectively performed a case-control study analyzing all patients with positive blood cultures (BCs) in the period of January 2017 to December 2017 at a 1100-bed teaching hospital in Italy and investigating the correlation between PCT and C-reactive protein (CRP) values (± 24 h from BC collection) in pathogens causing bloodstream infections. The study flowchart is presented in Additional file 1: Figure S1. During the study period, 1296 positive BCs were retrieved; of these, 258 (19.9%) episodes were recorded in the intensive care unit (ICU) and were included in the study. Moreover, 213 patients hospitalized in ICU with negative BC were used as control. Finally, 471 ICU patients were analyzed. Clinical characteristic and outcome of patients, according to BC results, are reported in Additional file 1: Table S1. As reported in Fig. 1, PCT concentrations (in nanograms per milliliter) were 25.1 ± 19.9 in patients with Gram-negative (GN) etiology, 29.9 ± 13.2 for Enterobacteriaceae, 8.9 ± 7.5 for Gram-positive (GP), and 2.1 ± 1.8 for fungi. Finally, in Additional file 1: Figure S1, receiver operating characteristic curves showed an area under the curve of 0.7 (95% confidence interval (CI) 0.62-0.77, P < 0.001) for PCT and 0.45 (95% CI 0.37-0.54, P = 0.32) for CRP among GN isolates, 0.74 (95% CI 0.67-0.81, P <0.001) for PCT and 0.49 (95% CI 0.4-0.57, P = 0.82) for CRP among Enterobacteriaceae, 0.46 (95% CI 0.39-0.53, P = 0.38) for PCT and 0.41 (95% CI 0.33-0.48, P = 0.01) for CRP among GP isolates, and 0.64 (95% CI 0.46-0.83, P = 0.22) for PCT and 0.59 (95% CI 0.45-0.73, P = 0.43) for CRP among fungi. Finally, logistic regression analysis showed that PCT values of more than 0.5 ng/mL and more than 10 ng/mL were independently associated with BCs positive for Enterobacteriaceae.
Our data confirmed the previous observations about the role of PCT and CRP in predicting BC results in critically ill patients [4,5]. Of interest, CRP was not able to predict BC results, whereas PCT values correlated with GN bacteremia and, among them, specifically identified Enterobacteriaceae. High PCT values (> 10 ng/mL) were independently associated with Enterobacteriaceae isolation. Even with the limitation of a single-center experience, these results might be useful to determine another role for PCT, helping physicians in the rapid identification of bacteremic ICU patients at risk of GN infection (especially Enterobacteriaceae) and driving the choice of a more appropriate empirical therapy.

Additional file
Additional file 1: Figure S1. Availability of data and materials Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
Authors' contributions MB, AR, ER, ED, and MM carried out the data collection and drafted the manuscript. NC, FD, AS, and FC participated in the design of the study and performed the statistical analysis. All authors read and approved the final manuscript.
Ethics approval and consent to participate Approved by local ethics review committee.

Consent for publication
Yes.