Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Observational Study
. 2019 Dec 10;14(12):e0225765.
doi: 10.1371/journal.pone.0225765. eCollection 2019.

Serum procalcitonin as an independent diagnostic markers of bacteremia in febrile patients with hematologic malignancies

Affiliations
Observational Study

Serum procalcitonin as an independent diagnostic markers of bacteremia in febrile patients with hematologic malignancies

Mina Yang et al. PLoS One. .

Abstract

Background: Serum procalcitonin (PCT) and C-reactive protein (CRP) are biomarkers of infection. In patients with hematologic disorders with or without hematopoietic stem cell transplantation (HSCT), it is difficult to distinguish bloodstream infections from aseptic causes of febrile episodes. The objective of this study was to investigate diagnostic values of PCT and CRP in predicting systemic bacterial infection in patients with hematologic malignancies.

Methods: Clinical and laboratory data of 614 febrile episode cases from 511 patients were analyzed. Febrile episodes were classified into four groups: (1) culture-positive bacterial infection by Gram-positive cocci (GPC), (2) culture-positive bacterial infection by Gram-negative bacilli (GNB), (3) fungal infection, and (4) viral infection or a noninfectious etiology.

Results: Of 614 febrile cases, systemic bacterial infections were confirmed in 99 (16.1%) febrile episodes, including 38 (6.2%) GPC and 61 (9.9%) GNB infections. PCT levels were significantly higher in GNB infectious episodes than those in febrile episodes caused by fungal infection (0.58 ng/mL (95% CI: 0.26-1.61) vs. 0.22 ng/mL (0.16-0.38), P = 0.047). Bacterial infectious episodes showed higher PCT and CRP levels than non-bacterial events (PCT: 0.49 (0.26-0.93) ng/mL vs. 0.20 (0.18-0.22) ng/mL, P < 0.001; CRP: 76.6 (50.5-92.8) mg/L vs. 58.0 (51.1-66.5) mg/L, P = 0.036). For non-neutropenic febrile episodes, both PCT and CRP discriminated bacteremia from non-bacteremia. However, in neutropenic febrile episodes, PCT only distinguished bacteremia from non-bacteremia. In non-neutropenic episode, both PCT and CRP showed good diagnostic accuracy (AUC: 0.757 vs. 0.763). In febrile neutropenia, only PCT discriminated bacteremia from non-bacterial infection (AUC: 0.624) whereas CRP could not detect bacteremia (AUC: 0.500, 95% CI: 0.439-0.561, P > 0.05).

Conclusions: In this single-center observational study, PCT was more valuable than CRP for discriminating between bacteremia and non-bacteremia independent of neutropenia or HSCT.

PubMed Disclaimer

Conflict of interest statement

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (NRF-2017R1A2B4011181), Republic of Korea. EONE Laboratories provided support in the form of salaries for YJ.K., but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. This commercial affiliation had no role with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc. This does not alter out adherence to PLOS ONE policies on sharing data and materials. All authors declare all potential competing interests for the purposes of transparency.

Figures

Fig 1
Fig 1
Comparison of PCT (left) and CRP (right) levels between bacteremia and non-bacteremia. Ninety-nine systemic bacterial infection episodes (colored box) showed higher PCT and CRP levels than nonbacterial events [median (95% CI) (PCT: 0.49 (0.26–0.93) ng/mL vs. 0.20 (0.18–0.22) ng/mL, P < 0.001; CRP: 76.6 (50.5–92.8) mg/L vs. 58.0 (51.1–66.5) mg/L, P = 0.036)] by Mann-Whitney U test.
Fig 2
Fig 2
Comparison of PCT (left) and CRP (right) levels between neutropenic and non-neutropenic patients. PCT levels discriminated bacteremia (gray box) from non-bacteremia (white box) in both neutropenia (-) patients [median (95% CI), 0.97 (0.27–5.61) ng/mL vs. 0.20 (0.17–0.24) ng/mL, P < 0.001] and neutropenia (+) patients [0.44 (0.19–0.78) ng/mL vs. 0.20 (0.18–0.22) ng/mL, P = 0.001]. CRP levels discriminated bacteremia (gray box) from non-bacteremia (white box) in neutropenia (-) patients [146.1 (78.9–240.2) mg/L vs. 52.2 (46.4–62.0) mg/L, P < 0.001)], but CRP levels were not different between bacteremia and non-bacteremia in neutropenia (+) patients [59.7 (41.7–84.5) mg/L vs. 69.5 (54.6–79.0) mg/L, P = 0.994] by Mann-Whitney U test.
Fig 3
Fig 3
Comparison of PCT (left) and CRP (right) levels between HSCT and non-HSCT. PCT levels discriminated bacteremia (gray box) from non-bacteremia (white box) in both non-HSCT patient [median (95% CI), 0.91 (0.27–5.51) ng/mL vs. 0.22 (0.18–0.28) ng/mL, P < 0.001] and HSCT patients [0.30 (0.18–0.58) ng/mL vs. 0.19 (0.16–0.22) ng/mL, P < 0.001]. CRP levels discriminated bacteremia (gray box) from non-bacteremia (white box) in non-HSCT patient [125.7 (91.3–159.8) mg/L vs. 67.4 (55.7–81.9) mg/L, P = 0.002] but CRP levels were not different between bacteremia and non-bacteremia in HSCT patients [51.0 (39.5–78.4) mg/L vs. 49.3 (39.9–61.3) mg/L, P = 0.414] by Mann-Whitney U test.
Fig 4
Fig 4
ROC curves of PCT and CRP for detecting bacteremia in total (a), non-neutropenic (b), and neutropenic (c) patients. In a total of 614 febrile episodes (a), the AUC of PCT was 0.651 (95% CI: 0.612–0.689) while that of CRP was 0.566 (95% CI: 0.526–0.606). In 341 non-neutropenic episodes (b), PCT and CRP showed AUC of 0.757 (95% CI: 0.708–0.801) and 0.763 (95% CI: 0.714–0.807), respectively. In 273 febrile neutropenia (c), PCT discriminated bacteremia from non-bacterial infection (AUC: 0.624, 95% CI: 0.564–0.682) whereas CRP could not detect bacteremia (AUC: 0.500, 95% CI: 0.439–0.561, P > 0.05).
Fig 5
Fig 5
Comparison of PCT (left) and CRP (right) levels among different etiologies of fever. PCT levels were significantly higher in GNB infectious episodes (gray box) than those in febrile episodes caused by fungal infection [median (95% CI), 0.58 (95% CI: 0.26–1.61) vs. 0.22 (0.16–0.38), P = 0.047] or other etiology [0.58 (95% CI: 0.26–1.61) vs. 0.19 (0.17–0.22), P < 0.001]. CRP levels were not significantly different among the four groups (P > 0.05).

Similar articles

Cited by

References

    1. Koya J, Nannya Y, Ichikawa M, Kurokawa M. The clinical role of procalcitonin in hematopoietic SCT. Bone Marrow Transplant. 2012;47(10):1326–31. Epub 2012/02/22. 10.1038/bmt.2012.18 . - DOI - PubMed
    1. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock, 2012. Intensive Care Medicine. 2013;39(2):165–228. 10.1007/s00134-012-2769-8 - DOI - PMC - PubMed
    1. Kim SC, Lee S, Kim S, Cho OH, Park H, Yu SM. Comparison of Clinical Performance Between BacT/Alert Virtuo and BacT/Alert 3D Blood Culture Systems. Annals of laboratory medicine. 2019;39(3):278–83. Epub 2019/01/10. 10.3343/alm.2019.39.3.278 - DOI - PMC - PubMed
    1. Cho SY, Park YJ, Cho H, Park DJ, Yu JK, Oak HC, et al. Comparison of Enterococcus faecium Bacteremic Isolates from Hematologic and Non-hematologic Patients: Differences in Antimicrobial Resistance and Molecular Characteristics. Annals of laboratory medicine. 2018;38(3):226–34. Epub 2018/02/06. 10.3343/alm.2018.38.3.226 - DOI - PMC - PubMed
    1. Kim DY, Lee Y-S, Ahn S, Chun YH, Lim KS. The Usefulness of Procalcitonin and C-Reactive Protein as Early Diagnostic Markers of Bacteremia in Cancer Patients with Febrile Neutropenia. Cancer Res Treat. 2011;43(3):176–80. 10.4143/crt.2011.43.3.176 - DOI - PMC - PubMed

Publication types

MeSH terms