Study location and patient
This study was conducted at the First Affiliated Hospital of Wenzhou Medical University, a 4100-bed teaching hospital (Wenzhou, China), over a 6-year period (1 January 2013 to 31 December 2018). Inclusion criteria of A. baumannii-induced bloodstream infection include[13]: (1) age ≥ 18 years; (2) blood culture positive for MDR-AB; (3) clinical signs consistent with infection; (4) the infection occurred ≥ 48 h after hospital admission. Only the first episode was included if the patient had multiple episodes of A. baumannii-induced BSI.
Study design and data collection
A retrospective study design was employed, with the main outcome measure being 30-day in-hospital mortality. Clinical information and laboratory parameters were collected from medical charts that used predefined definitions of variables. The collected data included demographic characteristics, underlying diseases, length of stay in hospital, treatments and procedures performed, laboratory results, antimicrobial therapies, and all-cause 30-day mortality.
Strains identification and antimicrobial susceptibility testing
The VITEK-2 automated system (BioMérieux, Marcy-l’Étoile, France) was used for isolates identification. Minimum inhibitory concentrations (MICs) of 14 antibiotics were tested by the agar dilution method, including imipenem, meropenem, ceftazidime, ceftriaxone, gentamicin, tobramycin, ciprofloxacin, levofloxacin, ampicillin/sulbactam, trimethoprim/sulfamethoxazole, piperacillin/tazobactam, piperacillin/tazobactam, cefperazone/sulbactam and tigecycline. The results were interpreted according to the recommendations of the Clinical and Laboratory Standards Institute (CLSI; Pittsburgh, PA, USA). The results of tigecycline were interpreted according to the recommendation of the Food and Drug Administration, with MICs of ≤2, 4, and ≥8 μg/ml interpreted as susceptible, intermediate, and resistant, respectively. MDR-AB were defined as non-susceptible to three or more different antimicrobial categories. All these strains were stored at - 80 °C for further research.
Multi-locus sequence typing(MLST)
In the present study, seven housekeeping genes (cpn60, fusA, gltA, pyrG, recA, rplB, and rpoB) were amplified and sequenced to characterize the genotypes of all isolates according to the provided protocols (www.pasteur.fr/mlst). The alleles and STs were assigned according to the online database of the Institut Pasteur’s MLST web site of A. baumannii.
Biofilm formation
The biofilm formation ability of the A. baumannii isolates on 96-well polystyrene microtiter plates was assessed using the crystal violet staining, as formerly described[14]. For evaluating biofilm formation, Muller-Hinton Broth and A. baumannii ATCC19606 were used as negative and positive controls, respectively. The ratio of the OD570 of the experimental isolate to the OD value of the negative control(ODC) was defined as the biofilm forming ability of the isolate. All the experiments were conducted in triplicate. The results were classified into the four following categories[15]: a) OD570 ≤ ODC = non-biofilm producer; b) ODC < OD570 ≤ 2ODC = weak biofilm producer; c) 2ODC < OD570 ≤ 4ODC = medium biofilm producer; d) 4ODC < OD570 = strong biofilm producer.
Serum resistance assay
Serum sensitivity assays were performed on clinical isolates[16]. 1 mL bacterial cultures were incubated until the bacterial suspension reached an OD600 of 0.5. Then washed and suspended in 1 mL of phosphate-buffered saline (PBS). Next, 100 μL of the bacterial suspension was mixed with 300 μL of normal human serum (NHS) and the mixture was incubated at 37°C for 3 h. Finally, calculate the colony-forming units (CFUs) of bacteria by single plate-serial dilution spotting (SP-SDS)[17]. The serum bactericidal effect was expressed as the ratio of the CFUs in the serum-bacteria suspension to the CFUs in a bacterial suspension without NHS. All experiments were performed in triplicate, and results were expressed as percent survival.
Galleria mellonella larva infection assay
The virulence of A. baumannii was estimated by infecting G. mellonella larvae, as described previously[8]. Briefly, a suspension of A. baumannii was prepared in PBS at approximately 1×108 CFU/mL, and 10 μL of this suspension (approximately 1×106 CFU) was injected through the last proleg of each larva. The control larvae were injected with 10 μL of PBS without bacteria. The survival rate of the G. mellonella was recorded for 144 hours. The larvae were considered dead when they were unresponsive to touch. 10 larvae were used for each experiment in triplicate
Statistical analysis
All statistical analyses were performed using SPSS 22.0 software (IBM, Armonk, NY, USA). The categorical variables were listed as percentages and the continuous data were expressed as mean± standard deviation (mean ± SD) or median (25th-75th percentile) appropriately. The Chi-squared test or the Fisher’s exact test was used for categorical variables, and the odds ratio (OR) was calculated with confidence intervals (CIs) of 95%. The continuous data were analyzed using the Student’s t test or Mann-Whitney U test. P-value < 0.05 was considered statistically significant. All tests were two-tailed. To determine the risk factors for A. baumannii-induced BSI, univariate logistic regression analyses were performed. All variables with a P-value <0.05 were included in the multivariate model.