| Objective:1 To investigate the bacterial distribution and drug resistance trends of infected patients in Wusong,branch of Zhongshan Hospital Affiliated to Fudan University,Shanghai from 2014 to 2019,and to provide decision-making basis for rational improvement of empirical antibiotic treatment in clinical practice.2 To explore risk factors and develop a prediction model for MDR-GNB infection in a comprehensive ICU,and to predict the risk of ICU patients acquiring MDR-GNB infection.Methods:1 We collected Bacteria isolated from infected patients in the comprehensive ICU of Wusong,branch of Zhongshan Hospital,Fudan University,Shanghai from2014 to 2019,and analyzed the bacterial spectrum and drug resistance of Gram-negative bacteria.2 We conducted a retrospective,matched(1:2)case-control study in comprehensive ICU in Wusong,branch of Zhongshan Hospital Affiliated to Fudan University in Shanghai from January 2014 to December 2019,the clinical data of patients were collected and divided into a case group and a control group according to whether or not MDR-GNB infection occurred,which were matched by the time of hospitalization.All data were randomly divided into training set and validation set by7:3.Univariate analysis was performed on the training set,and variables with univariate analysis results of P≤0.1 were included in multivariate logistic regression analysis to further identify independent risk factors for MDR-GNB infection.The risk factors based on multivariate logistic analysis were included in the nomogram,and the Bootstrap internal validation method was used to evaluate the nomogram.The discrimination and calibration of the model were evaluated by AUC,C-index and calibration curve,respectively.Plot the ROC curve of the validation set to validate the model.a=0.05.Result:1.We collected 594 strains of bacteria,including 341 strains of Gram-negative bacteria,accounting for 57.4% of the total strains,176 strains of Gram-positive bacteria,accounting for 29.6%,and 73 strains of fungi,accounting for 13%.The top five bacteria were Pseudomonas aeruginosa,accounting for 15.5%;Acinetobacter baumannii,accounting for 14.8%;Escherichia coli,accounting for 8.3%;Klebsiella pneumoniae,accounting for 8.1%;and Streptococcus,accounting for 8.1%.A total of157 strains of multi-drug-resistant gram-negative bacilli were collected,among which multi-drug resistant Acinetobacter baumannii was the main detected bacteria.From2014 to 2019,the number of multi-drug resistant gram-negative bacilli showed an overall upward trend.2.The drug resistance rates of Klebsiella pneumoniae to ampicillin/sulbactam,cefazolin,ceftriaxone,ceftazidime,cefepime,levofloxacin,aztreonam,ciprofloxacin,imipenem,gentamicin,tobramycin,cefotetan and co-trimoxazole were high,more than 50%,the drug resistance rates of Klebsiella pneumoniae to cefuroxime,cefoxitin,cefoperazone/sulbactam and nitrofurantoin were low,less than 30%.The resistance rates of multidrug-resistant Klebsiella pneumoniae to the 14 antibiotics studied were more than 50%;The resistance rates of Escherichia coli to ampicillin,ceftriaxone,Bactam,cefazolin,aztreonam,ciprofloxacin,levofloxacin were high,more than 50%,and the resistance rates of Escherichia coli to piperacillin/tazobactam,meropenem,cefuroxime,cefoxime,acetaminophen,cefotetan,imipenem,acamicin,tobramycin,cefoperazone/sulbactam and nitrofurantoin were all less than 30%,The resistance rates of MDR-E to 11 antibiotics were more than 50%;The resistance rates of Acinetobacter baumannii and multidrug-resistant Acinetobacter baumannii to Ampicillin/sulbactam,ceftazidime,cefepime,imipenem,ciprofloxacin,levofloxacin,piperacillin/tazobactam,gentamicin,meropenem were high,more than 50%,the resistance rate of Acinetobacter baumannii to cefoperazone/sulbactam was the lowest,which was 23.9%;The resistance rates of Pseudomonas aeruginosa to imipenem,ciprofloxacin and levofloxacin were 30%-50%,The resistance rates of Pseudomonas aeruginosa to Cefoperazone/sulbactam,acamicin,aztreonam,ceftazidime,tobramycin,gentamicin,meropenem,piperacillin/tazobactam,and cefepime were low,all below30%.The resistance rates of multidrug-resistant Pseudomonas aeruginosa to 7antibiotics were more than 30%.From 2014 to 2019,the resistance rates of Klebsiella pneumoniae to cefotetan,cefepime,imipenem,akamicin,gentamicin,ciprofloxacin and levofloxacin significantly increased;The resistance rates of E.coli to tobramycin,ciprofloxacin and levofloxacin increased;the resistance rate of Acinetobacter baumannii to cotrimoxazole decreased,from 66.7%-29%,Which to ceftazidime,cefepime,gentamicin and levofloxacin were on the rise,from(66.7%-77.8%)-(74.2-96.8);The resistance rate of Pseudomonas aeruginosa to imipenem increased,from 8.3% in 2014 to 54.5% in 2019,which to piperacillin/tazobactam,gentamicin and levofloxacin were on the decline.The difference was statistically significant(P<0.05).3.A total of 399 patients were included as study subjects.Among them,there were 241 males and 158 females,the ages were(24-99)years old,with an average age of(69.3±14.9)years old.Cardiovascular disease is the most common disease,accounting for 25.1%;followed by cerebrovascular disease,accounting for 12.3%;diabetes,accounting for 7.3%,and the mortality rate was 6.8%,lower respiratory tract infections and ventilator-associated pneumonia were the most common infections,and the mortality rate was 11.3% in the case group.Univariate analysis was performed in the training set,and the results showed that the case group and the control group had a history of gram-negative bacilli infection,types of comorbidities,diabetes,cardiovascular disease,cerebrovascular disease,hypoalbuminemia,and use of immunosuppressants,Apache II score,hospitalization days in ICU,days of indwelling central venous catheters,days of mechanical ventilation,days of indwelling urinary catheters,types of antibiotics used and days of antibiotics use were significantly different(P<0.05).Multivariate logistic analysis showed that cerebrovascular disease(OR=4.3,95%CI: 1.2-15),days of mechanical ventilation ≥7 days(OR=5.8,95%CI: 1.2-27.7),days of antibiotic use ≥3 days(OR=47,95%CI:10.4-212)was an independent risk factor for multidrug-resistant gram-negative bacilli infection in ICU.A nomogram model was established based on three factors.The calibration curve shows that the model has a good calibration.C-index value was0.933,and the AUC of training set and validation set was 0.955(95%CI: 0.925-0.984)and 0.898(95%CI: 0.836-0.961),respectively,indicating that the model had good predictive ability.Conclusion:1.Gram-negative bacteria are the main isolates in ICU in our hospital,and the detaction rate is on the rise.Gram-negative bacilli are highly resistant to a variety of antibiotics,the drug resistance of Klebsiella pneumoniae,Escherichia coli and Acinetobacter baumannii is even more severe,all of which have high resistance rates to quinolones and different degrees of resistance to other antibacterial drugs.The resistance rate of Pseudomonas aeruginosa to a variety of antibiotics decreased significantly.The resistance rates of MDR-GNB to most antibiotics were high.It is recommended to choose antibiotics reasonably according to the results of drug sensitivity in clinical treatment.2.The nomogram risk prediction model constructed according to cerebrovascular disease,mechanical ventilation ≥7 days,and antimicrobial use time≥3 days has high predictive value for multidrug-resistant gram-negative bacilli infection in the comprehensive ICU,which has high application value for guiding public health prevention and control and clinical work in hospitals and reducing the infection rate of multidrug-resistant gram-negative bacilli in ICU. |
PDF Full Text Request