Global Extraintestinal Pathogenic Escherichia coli (ExPEC) Lineages

Abstract

Extraintestinal pathogenic Escherichia coli (ExPEC) strains are responsible for a majority of human extraintestinal infections globally, resulting in enormous direct medical and social costs. ExPEC strains are comprised of many lineages, but only a subset is responsible for the vast majority of infections. Few systematic surveillance systems exist for ExPEC. To address this gap, we systematically reviewed and meta-analyzed 217 studies (1995 to 2018) that performed multilocus sequence typing or whole-genome sequencing to genotype E. coli recovered from extraintestinal infections or the gut. Twenty major ExPEC sequence types (STs) accounted for 85% of E. coli isolates from the included studies. ST131 was the most common ST from 2000 onwards, covering all geographic regions. Antimicrobial resistance-based isolate study inclusion criteria likely led to an overestimation and underestimation of some lineages. European and North American studies showed similar distributions of ExPEC STs, but Asian and African studies diverged. Epidemiology and population dynamics of ExPEC are complex; summary proportion for some STs varied over time (e.g., ST95), while other STs were constant (e.g., ST10). Persistence, adaptation, and predominance in the intestinal reservoir may drive ExPEC success. Systematic, unbiased tracking of predominant ExPEC lineages will direct research toward better treatment and prevention strategies for extraintestinal infections.

Keywords: Escherichia coli; extraintestinal infections; extraintestinal pathogenic E. coli; molecular epidemiology.

FIG 1

PRISMA flow diagram for this systematic review. This diagram shows the selection of studies included in the systematic review of ExPEC lineages (20).

FIG 2

Summary proportions from random-effects models for the most common ExPEC STs by study isolate inclusion criteria. Six studies were excluded from this analysis as they selected for isolates exhibiting specific drug resistance phenotypes, including amdinocillin and nitrofurantoin (1 study), nitrofurantoin alone (1 study), colistin (1 study), and fosfomycin (2 studies), or only for fully susceptible E. coli isolates (1 study). AMR, antimicrobial resistance.

FIG 3

Summary proportions from random-effects models for the most common ExPEC STs by geographic region. Six studies were not included: three included global collections of ExPEC, two studies were from Australia only, and one study did not report the geographic source of isolates analyzed.

FIG 4

Summary proportions from random-effects models for the most common ExPEC STs by source. Only those studies that examined isolates recovered exclusively from one source category were included. One study of bloodstream isolates also included isolates from orthopedic infections and was included in the bloodstream/meningitis group.

FIG 5

Summary proportions from random-effects models for the most common ExPEC STs by start date of study isolate collection. Dates were determined based on the reported start date for sample collection for each study. In some cases, isolate collection ended in a different period, which makes categories overlap for some studies. Moreover, the number of STs included in databases has grown over time. The absence of some STs might reflect the fact that these STs had not yet been added to the MLST allele database.