Association of microbiological factors with mortality in Escherichia coli bacteraemia presenting with sepsis/septic shock: a prospective cohort study

OBJECTIVES

This study aimed to determine the association of Escherichia coli microbiological factors with 30-day mortality in patients with bloodstream infection (BSI) presenting with a dysregulated response to infection (i.e. sepsis or septic shock).

METHODS

Whole-genome sequencing was performed on 224 E coli isolates of patients with sepsis/septic shock, from 22 Spanish hospitals. Phylogroup, sequence type, virulence, antibiotic resistance, and pathogenicity islands were assessed. A multivariable model for 30-day mortality including clinical and epidemiological variables was built, to which microbiological variables were hierarchically added. The predictive capacity of the models was estimated by the area under the receiver operating characteristic curve (AUROC) with 95% confidence intervals (CI).

RESULTS

Mortality at day 30 was 31% (69 patients). The clinical model for mortality included (adjusted OR; 95% CI) age (1.04; 1.02-1.07), Charlson index ≥3 (1.78; 0.95-3.32), urinary BSI source (0.30; 0.16-0.57), and active empirical treatment (0.36; 0.11-1.14) with an AUROC of 0.73 (95% CI, 0.67-0.80). Addition of microbiological factors selected clone ST95 (3.64; 0.94-14.04), eilA gene (2.62; 1.14-6.02), and astA gene (2.39; 0.87-6.59) as associated with mortality, with an AUROC of 0.76 (0.69-0.82).

DISCUSSION

Despite having a modest overall contribution, some microbiological factors were associated with increased odds of death and deserve to be studied as potential therapeutic or preventive targets.

Whole-genome characterisation of Escherichia coli isolates from patients with bacteraemia presenting with sepsis or septic shock in Spain: a multicentre cross-sectional study

BACKGROUND

Escherichia coli is the most frequent cause of bloodstream infections (BSIs). About one-third of patients with BSIs due to E coli develop sepsis or shock. The objective of this study is to characterise the microbiological features of E coli blood isolates causing sepsis or septic shock to provide exploratory information for future diagnostic, preventive, or therapeutic interventions.

METHODS

E coli blood isolates from a multicentre cross-sectional study of patients older than 14 years presenting with sepsis or septic shock (according to the Third International Consensus Definitions for Sepsis and Septic Shock criteria) from hospitals in Spain between Oct 4, 2016, and Oct 15, 2017, were studied by whole-genome sequencing. Phylogroups, sequence types (STs), serotype, FimH types, antimicrobial resistance (AMR) genes, pathogenicity islands, and virulence factors were identified. Susceptibility testing was performed by broth microdilution. The main outcome of this study was the characterisation of the E coli blood isolates in terms of population structure by phylogroups, groups (group 1: phylogroups B2, F, and G; group 2: A, B1, and C; group 3: D), and STs and distribution by geographical location and bloodstream infection source. Other outcomes were virulence score and prevalence of virulence-associated genes, pathogenicity islands, AMR, and AMR-associated genes. Frequencies were compared using χ² or Fisher's exact tests, and continuous variables using the Mann-Whitney test, with Bonferroni correction for multiple comparisons.

FINDINGS

We analysed 224 isolates: 140 isolates (63%) were included in phylogenetic group 1, 52 (23%) in group 2, and 32 (14%) in group 3. 85 STs were identified, with four comprising 44% (n=98) of the isolates: ST131 (38 [17%]), ST73 (25 [11%]), ST69 (23 [10%]), and ST95 (12 [5%]). No significant differences in phylogroup or ST distribution were found according to geographical areas or source of bloodstream infection, except for ST95, which was more frequent in urinary tract infections than in other sources (11 [9%] of 116 vs 1 [1%] of 108, p=0·0045). Median virulence score was higher in group 1 (median 25·0 [IQR 20·5-29·0) than in group 2 (median 14·5 [9·0-20·0]; p<0·0001) and group 3 (median 21 [16·5-23·0]; p<0·0001); prevalence of several pathogenicity islands was higher in group 1. No significant differences were found between phylogenetic groups in proportions of resistance to antibiotics. ST73 had higher median virulence score (32 [IQR 29-35]) than the other predominant clones (median range 21-28). Some virulence genes and pathogenicity islands were significantly associated with each ST. ST131 isolates had higher prevalence of AMR and a higher proportion of AMR genes, notably blaCTX-M-15 and blaOXA-1.

INTERPRETATION

In this exploratory study, the population structure of E coli causing sepsis or shock was similar to previous studies that included all bacteraemic isolates. Virulence genes, pathogenicity islands, and AMR genes were not randomly distributed among phylogroups or STs. These results provide a comprehensive characterisation of invasive E coli isolates causing severe response syndrome. Future studies are required to determine the contribution of these microbiological factors to severe clinical presentation and worse outcomes in patients with E coli bloodstream infection.

FUNDING

Instituto de Salud Carlos III.

Risk Factors and Predictive Score for Bacteremic Biliary Tract Infections Due to Enterococcus faecalis and Enterococcus faecium: a Multicenter Cohort Study from the PROBAC Project

Biliary-tract bloodstream infections (BT-BSI) caused by Enterococcus faecalis and E. faecium are associated with inappropriate empirical treatment and worse outcomes compared to other etiologies. The objective of this study was to investigate the risk factors for enterococcal BT-BSI. Patients with BT-BSI from the PROBAC cohort, including consecutive patients with BSI in 26 Spanish hospitals between October 2016 and March 2017, were selected; episodes caused by E. faecalis or E. faecium and other causes were compared. Independent predictors for enterococci were identified by logistic regression, and a predictive score was developed. Eight hundred fifty episodes of BT-BSI were included; 73 (8.5%) were due to target Enterococcus spp. (48 [66%] were E. faecium and 25 [34%] E. faecalis). By multivariate analysis, the variables independently associated with Enterococcus spp. were (OR; 95% confidence interval): cholangiocarcinoma (4.48;1.32 to 15.25), hospital acquisition (3.58;2.11 to 6.07), use of carbapenems in the previous month (3.35;1.45 to 7.78), biliary prosthesis (2.19;1.24 to 3.90), and moderate or severe chronic kidney disease (1.55;1.07 to 2.26). The AUC of the model was 0.74 [95% CI0.67 to 0.80]. A score was developed, with 7, 6, 5, 4, and 2 points for these variables, respectively, with a negative predictive value of 95% for a score ≤ 6. A model, including cholangiocarcinoma, biliary prosthesis, hospital acquisition, previous carbapenems, and chronic kidney disease showed moderate prediction ability for enterococcal BT-BSI. Although the score will need to be validated, this information may be useful for deciding empirical therapy in biliary tract infections when bacteremia is suspected. IMPORTANCE Biliary tract infections are frequent, and a significant cause of morbidity and mortality. Bacteremia is common in these infections, particularly in the elderly and patients with cancer. Inappropriate empirical treatment has been associated with increased risk of mortality in bacteremic cholangitis, and the probability of receiving inactive empirical treatment is higher in episodes caused by enterococci. This is because many of the antimicrobial agents recommended in guidelines for biliary tract infections lack activity against these organisms. To the best of our knowledge, this is the first study analyzing the predictive factors for enterococcal BT-BSI and deriving a predictive score.