Clinical analysis of distribution and drug resistance of pathogenic bacteria in blood culture of Dalian Municipal Central Hospital from 2015 to 2019

Objectives: To analyze the distribution of common pathogenic bacteria and pattern of drug resistance in the blood culture of inpatients. Methods: This was a descriptive study. Blood culture data of inpatients of Dalian Municipal Central Hospital from January 2017 to December 2020 were collected from microbiology laboratory for retrospective analysis. Results: A total of 24,786 specimens were submitted for examination from inpatients from 2015 to 2019, and 2131 strains of clinically non-repetitive pathogenic bacteria were identified. There were 1135 G-positive cocci (53.26%), including 248 strains of Staphylococcus hominis (21.85%) and 68 strains of Streptococcus species (5.99%). Other G-positive cocci 8 strains (0.70%). G-positive cocci were most sensitive to datomycin, linezolid and vancomycin. There were 923 G-negative bacilli (43.31%), including 476 strains (51.57%) of Escherichia coli, 244 strains (26.44%) of Klebsiella pneumoniae and 130 strains (14.08%) of Acinetobacter baumannii. G-negative bacilli were most sensitive to amikacin. Most of the blood specimens were obtained from the ICU patients (42.98%) followed by nephrology (8.68%) and respiratory medicine (7.32%). Conclusion: G-positive bacteria were mainly detected in the positive blood culture samples of inpatients in this hospital. Daptomycin, linezolid and vancomycin were preferred for G-positive cocci, while amikacin was highly sensitive to G-negative bacilli.


INTRODUCTION
Bloodstream infection is a serious systemic infectious disease, which refers to pathogenic microorganisms invade the blood circulation, reproduce in the blood, produce and release toxins and other metabolites, induce the release of cytokines, thus causing systemic infection, poisoning and systemic inflammatory reaction. 1 As the gold standard for diagnosing bloodstream infection, blood culture can provide accurate and reliable basis for clinical diagnosis of bloodstream infection and rational use of antibiotics. 2 Antimicrobial agents are the preferred treatment for infectious diseases whose extensive development and wide application will lead to their own gradually reduced susceptibility to pathogens. 3 Antimicrobial resistance (AMR) has an increasing impact on clinical treatment and has become a worldwide public health problem. 4 Studies have shown that inappropriate empiric antimicrobial therapy is an independent risk factor for increased mortality, mostly in patients with blood infections caused by Staphylococcus aureus or enterobacter. 5 Furthermore, different diseases may differ greatly in their pathogenic bacteria and drug resistance due to region and time, which in turn affects the treatment outcome. 6 In this study, a retrospective analysis was conducted on the principal pathogenic bacterial species and their drug resistance patterns in blood culture of inpatients in Dalian Municipal Central Hospital from 2015 to 2019, to provide a reference for clinicians in the use of antibiotics and the formulation of refractory infection control plans. All blood culture results from January 2015 to December 2019 were retrieved from the pathogenic microbiology Laboratory system database of our hospital. The specimen collection process during this period was as follows: Venous blood specimens from both sides of patients with systemic infection were collected according to aseptic procedures and injected into aerobic and anaerobic bottles respectively. After full mixing, the culture bottles were placed in the automatic blood culture instrument. Specimens that did not report positive results after six days of culture were judged to be negative. In case of a positive alarm of aerobic bacteria, the specimens should be transferred to sterile growth in time and cultured in a 35 0 C 5% CO 2 culture environment. In case of a positive alarm of anaerobe bacteria, the specimens should be transferred to sterile growth and put into anaerobic bags, and then cultured in an anaerobic environment. Strains were identified by Gram staining and VITEK 2 Compact automatic microbial analysis system was utilized to perform strain identification and drug susceptibility test in vitro on the isolated single colonies. All operations were carried out in strict accordance with the requirements of the 4th edition of the "National Guide to Clinical Laboratory Procedures", and the results are judged by referring to the M100-S31 document of the Clinical and Laboratory Standards Institute (CLSI). 7 Statistical Analysis: WHONT2020 software was used for drug sensitivity analysis, and the data were exported to calculate the ratio in an Excel list.

RESULTS
A total of 24,786 clinical specimens were submitted for examination from inpatients from January 2015 to December 2019, and 2131 strains of clinically non-repetitive pathogens were identified with a positive detection rate of 7.16%. The highest positive rate of pathogenic bacteria was found from the samples obtained in EICU (Emergency Intensive Care Unit), where 407 non-repeated pathogens were detected in 308 positive samples, accounting for 19.1% of the total detected strains and it was followed by MICU (Adult Intensive Care Unit), where 289 non-repetitive pathogens were detected in 239 specimens, accounting for 13.56%.
Among the 2131 strains of clinically nonrepetitive pathogens detected. There were 1135 G-positive cocci (     staphylococcus was significantly lower in 2017-2019 than in 2015 and 2016 (Table-II and Table-III). Among 923 strains of G-negative bacilli, Amikacin showed favorable antibacterial activity against Escherichia coli, Klebsiella pneumoniae, Enterobacter cloacae and Pseudomonas aeruginosa, with a sensitivity rate of >90%. Carbapenem antibacterial drugs had the optimal antibacterial activity against Escherichia coli, with a resistance rate of <2%, and also had high antibacterial activity against Klebsiella pneumoniae, with a drug resistance rate of < 10%; Enterobacter cloacae was not sensitive to carbapenems, with a drug resistance rate as high as 100%. In addition to amikacin, cefoperazone combined with sulbactam showed good antibacterial activity, with a sensitivity rate of > 80%; Pseudomonas aeruginosa was also insensitive to carbapenems, with a sensitivity of > 90% to tobramycin. Acinetobacter baumannii was highly sensitive to minocycline (84%) but insensitive to other commonly used antibacterial drugs (< 25%) ( Table-IV).

Bloodstream infection (BSI) is a systemic
infectious disease caused by the invasion of pathogenic microorganisms such as bacteria and fungi into the blood stream. 8 In recent years, with the continuous increase of invasive operations and irrational use of broad-spectrum antibiotics and corticosteroids, the incidence and mortality of BSI have been increasing year by year. 9 About 31 million cases of sepsis occur globally each year. 10 At present, blood culture is still the gold standard for diagnosing bloodstream infection. A total of 24,786 specimens were collected in this study, and 2131 strains of pathogenic bacteria were isolated and cultured, with a positive rate of 7.15%. G-positive cocci were the main positive strains, contrary to the results of Mao S et al. 11 , which may be related to the difference in specimen quantity, geographical location, and living habits. However, the results of this study were similar to Meidrops K et al. 12 , which may be related to the difference in the distribution of pathogenic bacteria in different departments.
Antibiotics are the first choice for clinical treatment of bloodstream infection, but because of the abuse of antibiotics, the drug resistance of pathogenic bacteria is increasingly serious. 13 Therefore, it is of great significance to carry out blood culture and drug sensitivity tests of pathogenic bacteria in early-stage to clarify the distribution and drug resistance of pathogenic bacteria to improve the rationality of drug use and control the development of the disease. 14 In this study, 407 positive strains were detected in EICU, accounting for 19.1% of the total positive strains, followed by MICU, with a total of 289 strains (13.56%) detected. Among the pathogenic bacteria detected in EICU, 57 strains of Staphylococcus epidermidis were highly resistant to penicillin G. No strains resistant to daptomycin, linezolid, and vancomycin were detected, which was similar to the reported drug resistance of Staphylococcus epidermidis. [15][16][17] A total of 317 strains of Escherichia coli were detected in the positive specimens, accounting for 14.88% of the total positive strains. The positive specimens were mainly concentrated in MICU (37 strains), EICU (34 strains), and Internal Hematology (32 strains), with a sensitivity rate of > 99% to imipenem and meropenem. This may be because medical patients mainly involve urinary tract infections or digestive tract infections. In patients with complex urinary tract infections, meropenem has been shown to have an advantage at the compound end point of clinical cure or improvement and microbial eradication. 18 A total of 206 pathogenic bacteria were detected in Surgery, mainly Escherichia coli (54 strains) and Klebsiella pneumoniae (33 strains), with multi-drug resistance, which was similar to the results in the literature. 19 Hepatobiliary Surgery detected the most positive strains (39 strains), while General Surgery detected only two positive strains, both of which were Klebsiella pneumoniae. Surgical inpatients were mostly surgery patients, so the prophylactic utilization of antibiotics may be the reason for the low positive detection rate and infection rate. 20 In this study, the main pathogens detected in the medical specimens were Escherichia coli. The most pathogenic bacteria were detected in Nephrology (185 strains), mainly Enterobacter cloacae (44 strains), followed by Respiratory Medicine (156 strains), mainly Staphylococcus hominis (43 strains). Significant differences were found in the number of positive detections and main pathogenic bacteria detected among the departments.

Limitations of this study:
We only analyzed and discussed the cases included in our hospital, which may not be representative enough. We look forward to a multi-center study in the future to reach more comprehensive conclusions.

CONCLUSION
G-positive bacteria were predominant in blood culture surface specimens of our hospital. Daptomycin, linezolid, and vancomycin are preferred for the treatment of G-positive cocci infections, and amikacin is highly sensitive in the treatment of G-negative bacilli. It is of great significance to pay close attention to the flora distribution, drug resistance and drug resistance change of bloodstream infection, guide the rational use of antibiotics and early control of infection, so as to prevent and control the outbreak of drugresistant strains.