Genomic surveillance reveals different transmission patterns between third-generation cephalosporin and carbapenem resistance in Klebsiella pneumoniae in the Comunidad Valenciana (Spain), 2018-2020

Abstract

Background: The emergence and spread of third-generation cephalosporins (3GC) and carbapenem-resistant Klebsiella pneumoniae pose a global critical challenge. Understanding the transmission dynamics within and between hospital environments is crucial to develop effective control strategies.

Methods: From 2017 to 2019, we conducted a genomic surveillance program in eight hospitals of the Comunitat Valenciana, Spain, collecting and sequencing 1,768 3GC- and carbapenem-resistant isolates. We quantified the overall transmission using core genomes and assessed the contribution of national and global isolates to the spread of AMR in the region by including 11,967 database genomes in the analysis.

Results: The local collection was highly diverse, involving 188 lineages, including global high-risk clones such as ST307 and ST11, and 3GC and carbapenem resistance determinants. Half of the isolates were involved in transmission, with 70.5% occurring within hospitals.

Conclusions: Different transmission patterns characterized the spread of 3GC- and carbapenem resistance in the region. While inter-hospital transmission played a significant role in the spread of 3GC-resistance, this was only sporadic for carbapenem resistance. Moreover, the factors behind inter-hospital spread for each type of resistance differed: while 3GC-resistance likely disseminated between hospitals through intermediate steps, carbapenem resistance was driven by more direct transmission routes. The burden of national and global cases on the ongoing regional AMR dissemination was low. Moreover, we revealed the rapid expansion in the region and globally of lineage ST307 carrying the bla_CTX-M-15 gene, a main driver of local transmissions, providing a deeper understanding of the successful spread of this high-risk clone.

Keywords: Klebsiella pneumoniae; Carbapenem resistance; Genomic surveillance.

Fig. 1

A Surveillance of Klebsiella pneumoniae in the Comunitat Valenciana (SKPCV). Map illustrating the geographical distribution of hospitals participating in the SKPCV project and the total number of samples collected from each hospital. The map is color-coded based on population density. B Antimicrobial resistance (AMR) profiles of isolates from each hospital, categorized by resistance to third-generation cephalosporins (3GC-R), carbapenems (Carbapenem-R), or susceptibility to both. C Temporal distribution of SKPCV sample collection. NA: non- available dates

Fig. 2

A The SKPCV phylogeny. Maximum likelihood tree of the 1,604 genomes of the SKPCV project using the core-genome (90%) alignment (Supplementary Fig. 2). B Association between lineages and resistant types. A simplified version of the tree in (A) showing only those sequence types (STs) with at least five isolates. The bar plot depicts the total number of isolates within each sublineage, color-coded according to their antibiotic resistance profile: susceptible, resistant to third-generation cephalosporins, or carbapenems. The last column represents the odds ratio (OR) and its 95 percent confidence interval estimates for the association between each sublineage and resistance to either 3GC or carbapenems. Only statistically significant values (p-value < 0.005) determined by t-tests are shown. When confidence intervals (CI) could not be computed due to small sample sizes, the odds ratio (OR) is shown without CI

Fig. 3

A Co-occurrence of STs and AMR genes across different hospitals. Each dot represents an SKPCV isolate, colored according to its source hospital. The gene labels in the y-axis are color-coded to differentiate genes encoding extended-spectrum beta-lactamases (ESBLs) in dark blue, AmpC beta-lactamases (AMPCs) in light blue, and carbapenemases in red. B Genetic diversity within hospitals. The y-axis depicts nucleotide diversity measured by single nucleotide polymorphisms (SNPs) within each hospital for various sets: All isolates, the 3GC-resistant (3GC-R) population, 3GC-R population excluding ST307 isolates carrying bla_CTX-M-15 gene (3GC-R-NoST307), the carbapenem-resistant (Carba-R) population, and Carba-R population excluding ST307 isolates carrying bla_CTX-M-15 gene (Carba-R-NoST307). C Net between-hospital divergences. The y-axis represents net genetic divergence (Da) measured by SNPs for the same populations as in B

Fig. 4

Isolates in Transmission Groups. A Isolates included or not in Transmission Groups (TGs). B TG sizes. Distribution of the TGs size and the isolates involved. Groups are color-coded by Sequence Type (ST). C ST involved in TG and resistance in each hospital. Isolates categorized into TGs or not, and antimicrobial resistance patterns, within the different hospitals color-coded by ST

Fig. 5

A Proportion of Transmission Groups (TGs) in intra-hospital or inter-hospital transmission. B Single Nucleotide Polymorphisms (SNPs) between TG in intra-hospital or inter-hospital transmission. Distribution of pairwise SNPs between SKPCV isolates in TGs limited to a single hospital (intra-hospital) or involving inter-hospital transmission, dividing the latter category into isolates collected in the same hospital (within-inter-hospital) and in different hospitals (between-inter-hospital). C TGs shared by hospitals. D Percentage of isolates within inter-hospital TG shared between hospitals. Each point represents a TG, with the y-axis representing the main hospital and the x-axis indicating the percentage of samples in the TG attributed to that hospital. The size of the dot reflects the size of the TG, the shape of the dots represents the resistant type, and dots are color-coded by Sequence Type (ST). E Global isolates included in SKPCV TGs. Local (Comunitat Valenciana), national (Spain), and global genomes included in SKPCV TGs as the SNP threshold for defining TGs increases

Fig. 6

Global phylogeny of lineage ST307. Maximum likelihood phylogenetic tree of ST307 genomes (n = 1,110). The core genome at 90% had 4,393 genes and resulted in an alignment of 4,268,326 bp with 16,936 variant positions (~ 3,967 SNPs/Mbp). Clades containing SKPCV isolates are highlighted in gray. Dots at the end of the tips are colored by the hospital where they were collected. The inner ring depicts the geographic origin. The second ring marks the presence or absence of the bla_CTX-M-15 and the third the coverage of the pCTX-M-15 assembled in this work. The fourth is the presence of bla_OXA-48 whereas the outermost ring is the coverage of the pOXA-48 assembled in this work