Characteristics and Clinical Implications of Carbapenemase-Producing Klebsiella pneumoniae Colonization and Infection, Italy

Abstract

Klebsiella pneumoniae carbapenemase–producing K. pneumoniae (KPC-Kp) has been endemic in Italy since 2013. In a multicenter cohort study, we investigated various aspects of KPC-Kp among patients, including 15-day mortality rates and delays in adequate therapy. Most (77%) KPC-Kp strains were sequence type (ST) ST512 or ST307. During 2017, KPC-Kp prevalence was 3.26 cases/1,000 hospitalized patients. Cumulative incidence of KPC-Kp acquired >48 hours after hospital admission was 0.68% but varied widely between centers. Among patients with mild infections and noninfected colonized patients, 15-day mortality rates were comparable, but rates were much higher among patients with severe infections. Delays of >4 days in receiving adequate therapy more frequently occurred among patients with mild infections than those with severe infections, and delays were less common for patients with known previous KPC-Kp colonization. Italy urgently needs a concerted surveillance system to control the spread of KPC-Kp.

Keywords: CRE; Carbapenem resistance; Enterobacteriaceae; Italy; KPC-Kp; Klebsiella pneumoniae; antimicrobial resistance; bacteria; healthcare-associated infections; mortality rates.

Figure 1

Flow chart of network of healthcare centers participating in a study of Klebsiella pneumoniae–carbapenemase producing K. pneumoniae (KPC-Kp), Italy, 2016–2018. The KPC-Kp network included 15 hospitals. Patients were included when KPC-Kp was diagnosed and excluded for various reasons. Hospitals were included when they submitted KPC-Kp–confirmed isolates and excluded from analysis when had no confirmed patients or did not enroll all confirmed patients. KPC-Kp, Klebsiella pneumoniae–carbapenemase producing K. pneumoniae.

Figure 2

Phylogenetic tree of 989 Klebsiella pneumoniae genomes isolated at hospitals participating in the KPC-producing K. pneumoniae (KPC-Kp) study, Italy. The key shows the number of isolates included in the study provided by each center; 2 samples (1 from each from hospitals A and I) were excluded because the total quality of the assemblies was not sufficient to have high confidence in the SNPs called through all the genome (total coverage <30). Inner circle shows the KPC-Kp mechanism identified; middle circle shows hospitals from which strains were isolated; and the outer circle the shows identified STs. The whole genome core single-nucleotide polymorphisms (SNPs) were extracted from the 989 K. pneumoniae genome assemblies by using kSNP3.0 (https://sourceforge.net/projects/ksnp). Parametric maximum-likelihood estimation (general time-reversible plus gamma distribution plus invariable sites) analysis with 1,000 bootstrap estimates was used to infer the phylogeny. We used IQ-TREE (http://www.iqtree.org) to generate the tree and iTOL (https://itol.embl.de) to draw the tree. Major STs are represented by branch colors; ST512 and ST307 were the predominant STs. Major branches have bootstrap values >0.75 for branch support. Scale bar indicates nucleotide substitutions per site. KPC, Klebsiella pneumoniae–carbapenemase; ST, sequence type.