The Lancet Microbe (May 2020)

Clinical relevance and plasmid dynamics of mcr-1-positive Escherichia coli in China: a multicentre case-control and molecular epidemiological study

  • Yan Jiang, PhD,
  • Ying Zhang, PhD,
  • Jun Lu, MS,
  • Qihui Wang, MS,
  • Yushan Cui, MS,
  • Yanfei Wang, MS,
  • Jingjing Quan, PhD,
  • Dongdong Zhao, MD,
  • Xiaoxing Du, MS,
  • Haiyang Liu, MS,
  • Xi Li, PhD,
  • Xueqing Wu, PhD,
  • Xiaoting Hua, PhD,
  • Ye Feng, PhD,
  • Yunsong Yu, ProfMD

Journal volume & issue
Vol. 1, no. 1
pp. e24 – e33

Abstract

Read online

Summary: Background: Although the emergence of the plasmid-mediated colistin resistance gene, mcr-1, caused global concern, little is known about its clinical implications and transmission characteristics over time. We aimed to investigate the clinical relevance of infection with mcr-1-positive Escherichia coli isolates and to investigate long-term plasmid dynamics. Methods: We did a multicentre case-control study and molecular epidemiological survey of mcr-1-positive E coli infections. E coli isolates from four hospitals in China in 2008 to 2017 were collected and screened for mcr-1 by PCR and Sanger sequencing. Patients with mcr-1-positive E coli infections and matched controls with mcr-1-negative E coli infections were included in a 1:4 ratio, considering age, sex, living environment, comorbidities, antimicrobials used, and clinical sample type as potential risk factors in a regression model. 28-day mortality was also observed. The genomes of all mcr-1-positive E coli were sequenced to explore their genetic background and map IS Apl1 elements. Plasmids carrying mcr-1 were characterised by their resistance profile and incompatibility group. Findings: 29 100 isolates were collected across all hospitals and during the study period, 300 (1·03%) of which were mcr-1-positive E coli. The overall proportion of mcr-1-positive isolates significantly increased from 0·42% (1 of 240) in 2008 to 1·39% (66 of 4748; p<0·0001) in 2017. Factors related to health-care contact, including receiving cancer care, indwelling central venous catheter, and hospitalisation in the past 3 months, and some sample types (pus secretion and sputum) were significantly associated with mcr-1-positive E coli infection. 28-day mortality was low in both mcr-1-positive (11 [4·4%] of 248 patients) and mcr-1-negative (39 [3·8%] of 1030 patients) groups and did not significantly differ. Although the genetic background of mcr-1-positive E coli was diverse, most of the mcr-1-encoding plasmids occurred in three dominant Inc groups (IncX4, IncI2, and IncHI2). Only the large IncHI2 plasmids conferred multiple resistances and probably integrated with other resistance plasmids. In 205 (68%) of 300 mcr-1-positive E coli isolates, mcr-1 genes lost their capacity for mobilisation because of loss of flanking IS Apl1 elements. Interpretation: The prevalence of mcr-1-positive E coli infection among patients increased over the study period, although it remained low. Health-care contact was the most probable risk factor. Plasmids are likely to have played a critical role in mcr-1 transmission, rather than clone dissemination and lateral transfer of IS Apl1. Our findings underscore the importance of continued surveillance of E coli strains positive for mcr-1 and potentially other resistance-associated genes, particularly in hospital settings. Funding: National Natural Science Foundation of China.