Carbapenemase-producing Klebsiella pneumoniae bloodstream infection in critically ill patients: risk factors and predictors of mortality.

A significant increase in carbapenemase-producing Klebsiella pneumoniae (CP-Kp) bacteraemias has been observed worldwide. The objective of the present work was to study the risk factors and predictors of mortality of CP-Kp bacteraemias among critically ill patients. During a 4-year period (2012-3015), a matched 1:2 case-control study was conducted. Klebsiella pneumoniae was identified by Vitek 2 technology. Antibiotic susceptibility was performed by the agar disc diffusion method and Etest. The presence of the bla KPC, bla VIM and bla NDM genes was confirmed by polymerase chain reaction (PCR). Epidemiologic data were collected from the intensive care unit (ICU) computerised database. One hundred and thirty-nine patients who developed a CP-Kp bacteraemia were matched with 278 patients. The majority of isolates (128; 92.1%) carried the bla KPC gene, seven carried both bla KPC and bla VIM, three bla VIM and one carried bla NDM. Risk factors for the development of CP-Kp bacteraemia were administration of tigecycline and number of antibiotics administered prior to CP-Kp bacteraemia. Overall, the 30-day mortality was 36.0%. Multivariate analysis revealed septic shock, Simplified Acute Physiology Score II (SAPS II) upon infection onset, adjunctive corticosteroid administration and parenteral nutrition as independent predictors of mortality, while treatment with a combination of appropriate antibiotics was identified as a predictor of good prognosis. Among septic shock patients (n = 74), Sequential Organ Failure Assessment (SOFA) score upon infection onset, adjunctive corticosteroid administration and strain carrying the bla KPC gene were independently associated with mortality, while the administration of combination treatment was identified as a predictor of a good prognosis. The administration of tigecycline predisposes to the induction of bacteraemia. Appropriate antibiotic treatment is associated with better survival, while concomitant corticosteroid treatment is associated with mortality.

Pitfalls in the identification of Enterococcus species and the detection of vanA and vanB genes.

UNLABELLED: The aims were to assess the performance of Vitek 2 in identifying enterococcal species and the implementation of GeneXpert(®) vanA/vanB PCR for the detection of vancomycin-resistant enterococci (VRE). Gram-positive cocci from clinical and environmental specimens (n = 431) suspicious of being enterococci by conventional methods were evaluated by Vitek 2. This system identified 296 Enterococcus faecium, 87 Enterococcus faecalis, 10 Enterococcus villorum, 9 Enterococcus gallinarum, 9 Enterococcus durans, 5 Enterococcus casseliflavus, 1 Enterococcus spp. and 14 isolates as Non-Enterococcus. All strains were submitted to pulsed field gel electrophoresis (PFGE) analysis showing 64 banding patterns. Representative strains from each banding pattern were further characterized to species level by 16S rDNA sequencing. The misidentification rate by Vitek 2 to species level among 429 molecularly identified enterococci was 6% (26 isolates). Additionally, 372 rectal swabs were obtained from critically ill patients. They were evaluated for the presence of VRE by ChromID VRE combined with in-house PCR vs GeneXpert(®) . GeneXpert(®) showed high (>92%) sensitivity, specificity, accuracy for vanA-positive Enterococcus detection, as well as, sensitivity and specificity for vanB-positive strains. Positive predictive value for detection of vanB-positive enterococci by GeneXpert(®) vanA/vanB was low (30%). GeneXpert(®) showed the same efficacy as ChromID VRE in detecting vanA-positive enterococci, but lower for vanB-gene detection. SIGNIFICANCE AND IMPACT OF THE STUDY: The study shows that even though the performance of Vitek 2 Advanced Expert System was good in identifying enterococci to species level, it is important to verify results by a molecular method when phenotypic findings are discordant with epidemiologic patterns. Furthermore, GeneXpert(®) vanA/vanB PCR and ChromID VRE combined with in-house PCR were applied in rectal samples for the detection of VRE colonization among critically ill patients. GeneXpert(®) showed an excellent performance in detecting vanA-positive enterococci, but false-positive results for vanB-gene detection render its application problematic in departments with high incidence of vanB-positive enterococci.

Number of positive blood cultures, biofilm formation, and adhesin genes in differentiating true coagulase-negative staphylococci bacteremia from contamination.

The significance of the number of coagulase-negative staphylococci (CNS)-positive blood cultures remains obscure in regards to determining true bacteremia versus contamination. The goal of this study was to determine the predictors of real CNS bloodstream infection among intensive care unit (ICU) patients. ICU patients with at least one CNS-positive blood culture were identified from the microbiology database. Biofilm formation was tested by glass tube and microtiter plate assay. mecA gene, ica operon genes (icaA, icaB, icaD), and adhesin genes (aap, bap, atlE, fbe, fnbA) were detected by polymerase chain reaction (PCR). CNS were recovered from 120 septic episodes, 20 of which were true CNS bacteremias, whereas from the remaining 100 episodes, the isolated CNS were characterized as contaminants. The number of positive blood cultures was significantly associated with true CNS bacteremia. Nineteen true bacteremic Staphylococcus epidermidis strains were compared to 38 contaminants. Biofilm synthesis was documented in 37 isolates associated with the presence of the ica operon (p = 0.048). There were 39, 26, 38, 21, and 10 strains positive for the presence of atlE, bap, fbe, aap, and fnbA genes, respectively. Rifampicin resistance, absence of severe sepsis, number of S. epidermidis-positive blood cultures, and absence of the bap gene were independently associated with true S. epidermidis bacteremia as compared to contaminant strains. The number of positive blood cultures is associated with true CNS bacteremia. The presence of adhesin genes may play a role in differentiating true infection from contamination, whereas absence of the bap gene is associated with true S. epidermidis bacteremia.