Background: Procalcitonin (PCT) is routinely used for an early recognition of severe infections and for promoting appropriate use of antibiotics. However, limited data correlating values of PCT with etiology of infection has been reported.Methods: During 2016, all positive blood cultures (BC) were retrospectively extracted in a 1100-beds Italian tertiary-care hospital. PCT and C-reactive protein (CRP) values were recorded within 24 h from BC collection. Primary endpoint of the study was to investigate the correlation between PCT and CRP values and the occurrence of bloodstream infections (BSI) caused by bacteria or fungi.Results: During the study period, 1296 positive BC were included: 712 (54.9%) due to Gram-positive (GP), 525 (40.5%) due to Gram-negative (GN) strains, and 59 (4.6%) caused by fungi. Among GN isolates, enterobacteriaceae were reported in 453 (86.3%) cases. PCT values were higher in patients with GN etiology (26.1 +/- 14.2 ng/mL) compared to GP (6.9 +/- 4.5) and fungi (3.3 +/- 2.4). Mean values for CRP in GN, GP, and fungi were not different. Receiver Operating Characteristic (ROC) curves showed an area under curve (AUC) of 0.71 for PCT and 0.51 for CRP among GN isolates; an AUC of 0.7 for PCT and 0.52 for CRP among enterobacteriaceae. Lower AUC for PCT were reported for GP and fungi.Conclusions: PCT showed moderate performance in early detection (within 24 h) of Gram-negative infections, especially those caused by enterobacteriaceae. Further prospective studies are mandatory to confirm these observations. (C) 2019 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University for Health Sciences.

Role of procalcitonin in predicting etiology in bacteremic patients: Report from a large single-center experience

Elda Righi;
2020-01-01

Abstract

Background: Procalcitonin (PCT) is routinely used for an early recognition of severe infections and for promoting appropriate use of antibiotics. However, limited data correlating values of PCT with etiology of infection has been reported.Methods: During 2016, all positive blood cultures (BC) were retrospectively extracted in a 1100-beds Italian tertiary-care hospital. PCT and C-reactive protein (CRP) values were recorded within 24 h from BC collection. Primary endpoint of the study was to investigate the correlation between PCT and CRP values and the occurrence of bloodstream infections (BSI) caused by bacteria or fungi.Results: During the study period, 1296 positive BC were included: 712 (54.9%) due to Gram-positive (GP), 525 (40.5%) due to Gram-negative (GN) strains, and 59 (4.6%) caused by fungi. Among GN isolates, enterobacteriaceae were reported in 453 (86.3%) cases. PCT values were higher in patients with GN etiology (26.1 +/- 14.2 ng/mL) compared to GP (6.9 +/- 4.5) and fungi (3.3 +/- 2.4). Mean values for CRP in GN, GP, and fungi were not different. Receiver Operating Characteristic (ROC) curves showed an area under curve (AUC) of 0.71 for PCT and 0.51 for CRP among GN isolates; an AUC of 0.7 for PCT and 0.52 for CRP among enterobacteriaceae. Lower AUC for PCT were reported for GP and fungi.Conclusions: PCT showed moderate performance in early detection (within 24 h) of Gram-negative infections, especially those caused by enterobacteriaceae. Further prospective studies are mandatory to confirm these observations. (C) 2019 The Authors. Published by Elsevier Limited on behalf of King Saud Bin Abdulaziz University for Health Sciences.
2020
Bacteremia
C-reactive protein
Enterobacteriaceae
Gram-negative
Procalcitonin
Adult
Bacteremia
Biomarkers
Female
Gram-Negative Bacterial Infections
Gram-Positive Bacterial Infections
Hospitalization
Humans
Intensive Care Units
Italy
Procalcitonin
ROC Curve
Retrospective Studies
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11562/1065346
Citazioni
  • ???jsp.display-item.citation.pmc??? 15
  • Scopus 21
  • ???jsp.display-item.citation.isi??? 20
social impact