Simple indictor of increased blood culture contamination rate by detection of coagulase-negative staphylococci

• Published in: Scientific Reports Vol. 11; no. 1; pp. 17538 - 5
• Main Authors: Yamamoto, Kei, Mezaki, Kazuhisa, Ohmagari, Norio
• Format: Journal Article
• Language: English
• Published: London Springer Science and Business Media LLC 02.09.2021; Nature Publishing Group UK; Nature Publishing Group; Nature Portfolio
• Subjects: 631/326/107; 631/326/2521; Antimicrobial agents; Bacteremia; Bacteremia - diagnosis; Bacteremia - enzymology; Bacteremia - microbiology; Bacteremia - pathology; Bacteria; Blood; Blood Culture; Blood Culture - classification; Blood Culture - methods; Clinical aspects; Coagulase; Coagulase - metabolism; Consent; Contamination; Cross Infection; Cross Infection - microbiology; Feedback; Humanities and Social Sciences; Humans; Laboratories; Medicine; multidisciplinary; Q; R; Retrospective Studies; ROC Curve; Science; Science (multidisciplinary); Staphylococcal Infections; Staphylococcal Infections - diagnosis; Staphylococcal Infections - enzymology; Staphylococcal Infections - microbiology; Staphylococcal Infections - pathology; Staphylococcus; Staphylococcus - isolation & purification; Staphylococcus - pathogenicity
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-021-96997-y

Coagulase-negative staphylococci (CoNS) are the most frequent contaminating bacteria; therefore, we aimed to investigate an indicator of CoNS to predict the increase in blood culture contamination rate (ConR). We performed a retrospective study of selected patients, who underwent blood culture testing. Contamination was defined as the presence of either one of two or more sets of skin-resident bacteria, except for cases with a low likelihood of contamination based on clinical aspects. We calculated the monthly ConR [(total number of contaminated cases per month)/(total number of blood culture sets collected per month) × 100] and analysed the ConR prediction ability using the following four indicators: the number of CoNS-positive sets of blood cultures, cases with at least one CoNS-positive blood culture set, cases with only one CoNS-positive blood culture set, and cases of contamination by CoNS. Cases with CoNS-positive blood cultures correlated with ConR (r = 0.85). Although the area under the receiver operating characteristic curve for the number of cases with ConR ≥ 2.5 differed significantly from that of the number of cases contaminated by CoNS, the negative predictive value was high, reaching up to 95.5% (95% confidential interval 87.3–99.1). The number of CoNS-positive cases could help predict an increase in ConR ≥ 2.5.