Clinical factors associated with 30-day mortality among patients with ST2 A. baumannii-induced BSI
A total of 108 patients with A. baumannii-induced BSI were enrolled during the 6-year study period. Seventy-five patients died within 30 days, with a mortality rate of 69.4%. The demographics between the non-survivor group and survivor group were similar, as shown in Table 1. Briefly, the majority of patients in both groups were elderly and male patients. The most common comorbidity was pneumonia. MDR-AB was also isolated from sputum in 86 patients (79.6%) during the hospitalization. Empirical antimicrobial therapy was performed on all patients. The univariate regression analysis revealed that the use of mechanical ventilation (OR 2.49, 95%CI [1.00-6.22]), ICU stay (OR 8.82, 95%CI [3.38-23.02]), and thrombocytopenia (OR 8.72, 95%CI [1.93-39.30]) were all significant risk factors associated with patients mortality (Table 1). Furthermore, deep vein intubation and other microbial isolation were not significantly associated with survival (P=0.247, fungi P=0.890, and bacteria P =0.642, respectively). In contrast, change of antimicrobial agents within 48 h after the bacteria were isolated from blood, the use of the antibacterial agent combination and more inpatient days were significantly associated with survival (P <0.001, P =0.037, and P =0.007, respectively)(Table 1).
Antimicrobial treatment regimens and clinical outcomes of 108 patients are listed in Table 2. During hospitalization, the usage rate of carbapenem, cefoperazone/sulbactam, tigecycline, piperacillin/tazobactam and polymyxin were 85.2% (92/108), 69.4% (75/108), 45.4% (49/108), 27.8% (30/108) and 7.4% (8/108), respectively. However, the usage of these antimicrobials was not significantly associated with survival (P >0.05). Detailed antimicrobials therapy options after bacteria isolation are listed in Table 2.
The most common treatment option was the combination of cefoperazone/sulbactam and tigecycline (25.9%, 28/108), with 53.6% (15/28) survival rate, followed by the monotherapy of carbapenem, cefoperazone/sulbactam and tigecycline (14.8%, 13.0%, and 5.6%, respectively), and the survival rates were 25% (4/16) 28.8% (4/14) and 0% (0/6), respectively. Nine patients were treated by combining carbapenem and tigecycline; however, no patients survived. Among 5 patients treated by combining polymyxin and tigecycline, 2 patients survived.
Antimicrobial Susceptibility and MLST of A. baumannii
The resistance rates of 108 A. baumannii were the following: imipenem 100%, meropenem 100%, ceftriaxone 100%, ciprofloxacin 100%, ceftazidime 100%, ampicillin/sulbactam 99.1%, piperacillin/tazobactam 99.1%, gentamicin 98.2%, levofloxacin 94.4%, tobramycin 85.2%, trimethoprim/sulfamethoxazole 76.7%. The resistance rate, intermediate rate, and sensitive rate of cefperazone/sulbactam were 77.8%, 18.5% and 3.7%, and those of tigecycline were 6.5%, 5.5% and 88.0%, respectively(Table 3). Among 61 isolates (56.5%) presented an XDR phenotype, most were sensitive only to polymyxin and tigecycline. The results of MLST showed that all isolates were ST2.
Pathogenicity of A. baumannii
Thirty A. baumannii isolated from the survivors and 30 A. baumannii isolated from the non-survivors were randomly selected for biofilm formation experiment and serum resistance assay. Among those who survived, the percent of weak biofilm producer, medium biofilm producer, and strong biofilm producer were 13.3%, 66.7%, and 20%, respectively. Among those who died, the percent of weak biofilm producer, medium biofilm producer, and strong biofilm producer was 6.7%, 83.3%, and 10%, respectively. However, the biofilm formation ability in the two groups was not statistically different (3.36±1.30 vs 3.13±0.77; P =0.417) (Figure 1). On the other hand, after 3 h incubation with normal human serum, the survival rate of A. baumannii isolated from non-survivors was higher than A. baumannii isolated from survivors. Still, the survival rates were not statistically different among the two groups (17.5% [6.9% - 38.0%] vs. 10.5% [6.2% - 27.0%]; P =0.209((Figure 2). The survival rate of the standard strain ATCC19606 was less than 0.1%.
Furthermore, 10 A. baumannii isolated from the survivors and 10 A. baumannii isolated from the non-survivor were randomly selected for G. mellonella larva infection assay. G. mellonella survival data are shown in Figure 3. The survival rates of the infected larvae between the two groups were not statistically different in every observation between day 1 and day 6 (P =0.522). The pathogenicity of ST2 was significantly higher than ATCC 19606 (P <0.001).