Temporal dynamics of genetically heterogeneous extended-spectrum cephalosporin-resistant Escherichia coli bloodstream infections

Abstract

Extended-spectrum cephalosporin-resistant Escherichia coli (ESC-R- Ec ) is an urgent public health threat with sequence type clonal complex 131 (STc131), phylogroup B2 strains being particularly concerning as the dominant cause of ESC-R- Ec infections. To address the paucity of recent ESC-R- Ec molecular epidemiology data in the United States, we used whole-genome sequencing (WGS) to fully characterize a large cohort of invasive ESC-R- Ec at a tertiary care cancer center in Houston, Texas, collected from 2016 to 2020. During the study time frame, there were 1,154 index E. coli bloodstream infections (BSIs) of which 389 (33.7%) were ESC-R- Ec . Using time series analyses, we identified a temporal dynamic of ESC-R- Ec distinct from ESC-susceptible E. coli (ESC-S- Ec ), with cases peaking in the last 6 months of the calendar year. WGS of 297 ESC-R- Ec strains revealed that while STc131 strains accounted for ~45% of total BSIs, the proportion of STc131 strains remained stable across the study time frame with infection peaks driven by genetically heterogeneous ESC-R- Ec clonal complexes. bla CTX-M variants accounted for most β-lactamases conferring the ESC-R phenotype (89%; 220/248 index ESC-R -Ec ), and amplification of bla CTX-M genes was widely detected in ESC-R- Ec strains, particularly in carbapenem non-susceptible, recurrent BSI strains. Bla CTX-M-55 was significantly enriched within phylogroup A strains, and we identified bla CTX-M-55 plasmid-to-chromosome transmission occurring across non-B2 strains. Our data provide important information regarding the current molecular epidemiology of invasive ESC-R- Ec infections at a large tertiary care cancer center and provide novel insights into the genetic basis of observed temporal variability for these clinically important pathogens. IMPORTANCE Given that E. coli is the leading cause of worldwide ESC-R Enterobacterales infections, we sought to assess the current molecular epidemiology of ESC-R- Ec using a WGS analysis of many BSIs over a 5-year period. We identified fluctuating temporal dynamics of ESC-R- Ec infections, which have also recently been identified in other geographical regions such as Israel. Our WGS data allowed us to visualize the stable nature of STc131 over the study period and demonstrate a limited but genetically diverse group of ESC-R- Ec clonal complexes are detected during infection peaks. Additionally, we provide a widespread assessment of β-lactamase gene copy number in ESC-R- Ec infections and delineate mechanisms by which such amplifications are achieved in a diverse array of ESC-R- Ec strains. These data suggest that serious ESC-R- Ec infections are driven by a diverse array of strains in our cohort and impacted by environmental factors suggesting that community-based monitoring could inform novel preventative measures.

Identification of A Novel CG307 Sub-clade in Third Generation Cephalosporin ResistantKlebsiella pneumoniaeCausing Invasive Infections in the United States

“…Consistent with our observation 3GC-RKp infections tended to occur in the warmer second half of the calendar year compared to the first half of the year, a multi-site surveillance study across several continents found that environmental factors associated with warmer months may increase incidence rates of K. pneumoniae infections [58]. Nevertheless, similar to a recent analysis on 3GC-R E. coli prevalence at our institution [42], we observed a statistically significant association only for 3GC-RKp but not 3GC-SKp infections. Thus, our results taken together with previous studies suggest that limiting 3GC-RKp acquisition outside of the hospital is likely to be critical to mitigation efforts.…”

“…1A). Similar to our recent study of 3GC-R Escherichia coli bacteremia [42], 3GC-RKp bacteremia prevalence was significantly higher in the last six months of the year relative to the first six months (0.67 vs. 0.45 bacteremias per month per 1,000 patient admissions respectively; Student's t-test P-value < 0.001) (Fig. 1B, 1C).…”

Identification of a novel CG307 sub-clade in third-generation-cephalosporin-resistant Klebsiella pneumoniae causing invasive infections in the USA

“…Consistent with our observation, 3GC-R K. pneumoniae infections tended to occur in the warmer second half of the calendar year compared to the first half of the year, a multi-site surveillance study across several continents found that environmental factors associated with warmer months may increase incidence rates of K. pneumoniae infections [57]. Nevertheless, similar to a recent analysis on 3GC-R E. coli prevalence at our institution [43], we observed a statistically significant association only for 3GC-R K. pneumoniae but not 3GC-S K. pneumoniae infections. Thus, our results taken together with previous studies suggest that limiting 3GC-R K. pneumoniae acquisition outside of the hospital is likely to be critical to mitigation efforts.…”

“…When normalized by the number of admissions to account for fluctuations in patient volume, the rates of index 3GC-R K. pneumoniae bacteraemias were not significantly different over the course of the study (Mann–Kendall trend test, P value >0.05) (). Similar to our recent study of 3GC-R Escherichia coli bacteraemia [43], 3GC-R K. pneumoniae bacteraemia prevalence was significantly higher in the last 6 months of the year relative to the first 6 months (0.67 vs 0.45 bacteraemias per month per 1000 patient admissions, respectively; Student’s t -test, P value <0.001) (). No statistically significant difference in 3GC-S K. pneumoniae bacteraemias stratified by time of year was observed (1.6 vs 1.3 bacteraemias per month per 1000 patient admissions in the second half of year vs the first half of year, respectively; Students t -test, P value=0.07; ).…”

“…1a). Similar to our recent study of 3GC-R Escherichia coli bacteraemia [43], 3GC-R K. pneumoniae bacteraemia prevalence was significantly higher in the last 6 months of the year relative to the first 6 months (0.67 vs 0.45 bacteraemias per month per 1000 patient admissions, respectively; Student's t-test, P value <0.001) (Fig. 1b, c).…”