Increased multidrug resistant isolates: new clinical burdens for 66 hospitals in Shanghai, 2015 to 2017.

BACKGROUND

To investigate the etiology of urinary tract and respiratory infections and the rate of drug resistant bacteria in Shanghai from 2015 to 2017, provide the support to the promotion of the correct clinical etiological specimens detection.

METHODS

The retrospective questionnaire survey was conducted to obtain the antibiotic prescription rate, pathogen detection rate, and isolation rate of drug-resistant bacteria of the inpatients in 66 hospitals of Shanghai in 2015 and 2017.

RESULTS

Although the detection rate of multidrug-resistant bacteria in sputum specimens has decreased, the clinical pathogen detection still relied mainly on sputum specimens in 2017. Among the sputum specis, the detection rates of extended-spectrum β-lactamase-producing Escherichia (ESBL-ESC), extended-spectrum β-lactamase Klebsiella pneumoniae (ESBL-KP), carbapenem-resistant Acinetobacter baumannii (CRAB), carbapenem-resistant Pseudomonas aeruginosa (CRPA) and methicillin-resistant Staphylococcus aureus (MRSA) in 2017 were 66.67%, 32.46%, 61.74%, 32.01% and 58.55% respectively. The detection rates of ESBL-ESC, ESBL-KP, CRAB, CRPA in 2017 were increased while the MRSA was decreased than 2015 (P<0.001). Among the blood samples, the detection rates of ESBL-ESC, ESBL-KP, CRAB, CRPA, MRSA and vancomycin-resistant Enterococcus sp (VRE) in 2017 were 53.71%, 31.43%, 50.80%, 19.43%, 43.87% and 0.55% respectively. The detection rates of ESBL-KP, CRAB, CRPA and MRSA increased while the rates of ESBL-ESC and VRE decreased compared with 2015 (P<0.005). The pathogens of multi-drug resistant bacteria were mainly detected from sputum specimens in 2017, which were all higher than detected from the blood specimens (P<0.001).

CONCLUSIONS

Most of the multi-drug resistant bacteria in Shanghai, especially in Acinetobacter baumannii or Pseudomonas aeruginosa are mainly detected from sputum specimens, indicating that the actual drug resistance may be overestimated.