Objective: To analyze the distribution and drug resistance of pathogen isolated from severely burned patients with bloodstream infection, so as to provide reference for the clinical treatment of these patients.

Methods: Blood samples of 162 severely burned patients (including 120 patients with extremely severe burn) with bloodstream infection admitted into our burn ICU from January 2011 to December 2014 were collected. Pathogens were cultured by fully automatic blood culture system, and API bacteria identification panels were used to identify pathogen. Kirby-Bauer paper disk diffusion method was used to detect the drug resistance of major Gram-negative and -positive bacteria to 37 antibiotics including ampicillin, piperacillin and teicoplanin, etc. (resistance to vancomycin was detected by E test), and drug resistance of fungi to 5 antibiotics including voriconazole and amphotericin B, etc. Modified Hodge test was used to further identify imipenem and meropenem resistant Klebsiella pneumonia. D test was used to detect erythromycin-induced clindamycin resistant Staphylococcus aureus. The pathogen distribution and drug resistance rate were analyzed by WHONET 5.5. Mortality rate and infected pathogens of patients with extremely severe burn and patients with non-extremely severe burn were recorded. Data were processed with Wilcoxon rank sum test.

Results: (1) Totally 1 658 blood samples were collected during the four years, and 339 (20.4%) strains of pathogens were isolated. The isolation rate of Gram-negative bacteria, Gram-positive bacteria, and fungi were 68.4% (232/339), 24.5% (83/339), and 7.1% (24/339), respectively. The top three pathogens with isolation rate from high to low were Acinetobacter baumannii, Staphylococcus aureus, and Pseudomonas aeruginosa in turn. (2) Except for the low drug resistance rate to polymyxin B and minocycline, drug resistance rate of Acinetobacter baumannii to the other antibiotics were relatively high (81.0%-100.0%). Pseudomonas aeruginosa was sensitive to polymyxin B but highly resistant to other antibiotics (57.7%-100.0%). Enterobacter cloacae was sensitive to imipenem and meropenem, while its drug resistance rates to ciprofloxacin, levofloxacin, cefoperazone/sulbactam, cefepime, piperacillin/tazobactam were 25.0%-49.0%, and those to the other antibiotics were 66.7%-100.0%. Drug resistance rates of Klebsiella pneumoniae to cefoperazone/sulbactam, imipenem, and meropenem were low (5.9%-15.6%, two imipenem- and meropenem-resistant strains were identified by modified Hodge test), while its drug resistance rates to amoxicillin/clavulanic acid, piperacillin/tazobactam, cefepime, cefoxitin, amikacin, levofloxacin were 35.3%-47.1%, and those to the other antibiotics were 50.0%-100.0%. (3) Drug resistance rates of methicillin-resistant Staphylococcus aureus (MRSA) to most of the antibiotics were higher than those of the methicillin-sensitive Staphylococcus aureus (MSSA). MRSA was sensitive to linezolid, vancomycin, and teicoplanin, while its drug resistance rates to compound sulfamethoxazole, clindamycin, minocycline, and erythromycin were 5.3%-31.6%, and those to the other antibiotics were 81.6%-100.0%. Except for totally resistant to penicillin G and tetracycline, MSSA was sensitive to the other antibiotics. Fourteen Staphylococcus aureus strains were resistant to erythromycin-induced clindamycin. Enterococcus was sensitive to vancomycin and teicoplanin, while its drug resistance rates to linezolid, chloramphenicol, nitrofurantoin, and high unit gentamicin were low (10.0%-30.0%), and those to ciprofloxacin, erythromycin, minocycline, and ampicillin were high (60.0%-80.0%). Enterococcus was fully resistant to rifampicin. (4) Fungi was sensitive to amphotericin B, and drug resistance rates of fungi to voriconazole, fluconazole, itraconazole, and ketoconazole were 7.2%-12.5%. (5) The mortality of patients with extremely severe burn was higher than that of patients with non-extremely severe burn. The variety of infected pathogens in patients with extremely severe burn significantly outnumbered that in patients with non-extremely severe burn (Z=-2.985, P=0.005).

Conclusions: The variety of pathogen in severely burned patients with bloodstream infection is wide, with the main pathogens as Acinetobacter baumannii, Staphylococcus aureus, and Pseudomonas aeruginosa, and the drug resistance situation is grim. The types of infected pathogen in patients with extremely severe burn are more complex, and the mortality of these patients is higher when compared with that of patients with non-extremely severe burn.