| Objective:1.Understand the basic situation of the hospital infection of multi drug resistant bacteria in ICU.2.Analysis of the related factors of nosocomial infection of multi drug resistant bacteria.3.Logistic regression risk model of multiple drug resistant bacteria infection in ICU was evaluated.Methods:A retrospective study of patients who were hospitalized in a comprehensive ICU teaching hospital from September 2014 to October 2012. A retrospective study combine with paper/electronic medical records and hospital infection monitoring software database. 836 patients were investigated, Including the ICU patients>48 h and transfer the ICU < 48 h. All data were processed by SPSS16.0 statistical software. For single factor chi-square analysis testing data; the single factor analysis was statistically significant results in multiple factors analysis, for unconditional "gradually" Logistic regression to establish risk model; for unconditional Logistic regression to establish "gradually" risk model; test of regression model of total effectiveness, goodness of fit,discrimination ability, evaluation model of wide and predictive ability(Modeling the data from ICU2012.10 2014.9 in hospitalized patients, the reserved data from ICU2014.10-2015.9 hospitalized patients).Results:1.ICU hospital infection incidence of multi-resistant bacteria : In October 2012 to September 2015, 836 cases of hospitalized patients with investigations. There were 468 males and 368 females, the average infection rate of 14.23%. compared to the ICU hospital infection multi-resistant bacteria, statistically significant difference(X2 = 1.672′103, P< 0.0001).2.ICU patients catheter indwelling and associated infection : 836 Cases of ICU hospitalized days were 12788 d. Central venous catheter, urine tube, breathing machine using a number of days of 10522 d, 7891 d, 4775 d, lien rate were 82.30%, 61.71%and 37.34% on average. The average infection rate were 1.71%, 1.71% and 3.35%.Blood vessel related bloodstream infection rate is highest, followed by breathing machine intubation patients. The average of the three pipe infection rate was statistically difference(X2 = 19.334, P < 0.0001).3.Proportion of multiple drug resistant bacteria infection : Lower respiratory tract infections(53 cases, 24.42%) of the first; followed by bacteremia(41 cases, 18.89%);catheter-related bloodstream infection(28 cases, 12.90%); urinary tract infection(23cases, 10.60%); catheter-related infection(18 cases, 8.29%); ventilator-associated infection(16 cases, 7.37%); gastrointestinal tract infection(11 cases, 7.37%); organ lacunar infection(7 cases, 5.07%%); central nervous system infections(6 cases,2.76%); pleural infection(4 cases, 1.85%); incision infection(4 cases, 1.85%); Other parts of the infection(6 cases, 2.77%).4.The distribution of pathogenic bacteria of ICU specimens : 119 cases of ICU hospital infection cases multi-resistant bacteria, a total of 340 strains resistant strains.Dominated by Gram negative, the main pathogenic bacteria mainly as follows :Acinetobacter baumannii(95 strains, 27.94%) accounted for the first, followed by Klebsiella pneumoniae(51 strains, 15.00%), Enterococcus faecium(42 strains,12.35%); coagulase- negative staphylococci(41 strains, 12.06%); Copper Green Pseudomonas aeruginosa(24 strains, 7.06%); EOS malt narrow food unit cell(16strains, 4.71%); Enterobacter cloacae(16 strains, 4.71%). Pathogenic bacteria in sputum samples for the first,mainly acinetobacter baumannii and pneumonia klebsiella bacteria;Secondly from blood specimens of pathogenic bacteria(80 strains, 23.53%),mainly negative staphylococcus of solidification; Pathogenic bacteria in urine sample(31strains, 9.12%), mainly have excrement enterococcus; Center vein pipe specimens of pathogenic bacteria(27 strains, 7.94%) mainly solidification negative staphylococcus aureus and acinetobacter baumannii.5.ICU main pathogenic bacteria drug resistance situation: Acinetobacter baumannii Klebsiella pneumoniae, grams Klebsiella sp, Enterococcus faecium, coagulase negative staphylococci of cephalosporin antibiotic resistance is higher, which Pseudomonas aeruginosa of cephalosporin antibiotics is sensitive.6.ICU antibiotics usage and inspection rate: 2012.10-2015.9 antibiotic utilization rate of 85.93%, 84.59%, 90.82%, the average utilization rate of antibiotics was87.20%; One of the league, dual, triple, greater than or equal to quadruple medicine followed by an average of 410 cases, 180 cases, 103 cases, 36 cases, accounted for58.24%, 24.69%, 14.13%, 58.24%. Results show that the antibiotic use is still in the high level, with a united medication is given priority to, as the number of combinations is reduced, antibiotic usage also decreases.7.ICU hospital infection were multi-resistant bacteria mortality : ICU inpatients total mortality was 5.74%. Among them, in hospital infection were multi-resistant bacteria mortality rate 12.61%; No infection were multi-resistant bacteria mortality was4.61%8.Single factor analysis : through case-control study method, the multi drug resistant bacteria infection in 75 patients as the case group, without the occurrence of multi drug resistant bacteria infection in 467 cases as control group, to analyze the relationship between the factors associated with hospital ICU of multi drug resistant bacteria infection. Single factor analysis showed that: the number of combined use of antibiotics and the number of days, the basic disease, the use of three days and the number of times, etc, 17 factors are the risk factor of multiple drug resistant bacteria infection in hospital(the data of 2012.9-2014.10 was analyzed by chi square).9.Analysis of multi factors : multi factor results show ICU length of stay(OR=2.249 95%CI=1.473 ~ 3.4344), disease(OR=1.456 95%CI=1.132 ~ 1.874),intubation ventilator days(OR=1.514 95%CI=1.161 ~ 1.976), hypoproteinemia(OR=87.020 95%CI=27.988 ~ 270.568), antibiotics combined with(OR=1.19095%CI=1.019 ~ 1.391 for), fever(OR=10.536 95%CI=1.080 ~ 102.768) six factors are the independent risk factors of multi drug resistant bacteria infection in comprehensive ICU.Logistic regression equation : logistic(P) =-10.222+0.811X1(ICU length of stay)+0.376X2(basic disease) +0.415X3(using ventilator days) +4.466X4(hypoproteinemia)+0.452X5(antibacterial drugs in combination +2.355X6(feve).10.Evaluation of the logistic regression models : logistic regression model total effectiveness of likelihood ratio chi square(likelihood ratio chi square =142.264,DF=6, P < 0.0001; Wald test, Wald chi square(Wald chi-Square) =286.123, DF=6,P < 0.0001; the logistic regression equation and test of goodness of fit, the Hosmer lemeshow goodness of fit test method, Chi-Square=0.863, DF=8, P=0.999; the area under the ROC curve was 0.972, sensitivity 92.4%, specificity 91.1%; model predictive ability was 91.8%.Conclusions:1.ICU is a high incidence of multiple drug resistant bacteria in hospital department,should strengthen the comprehensive ICU of the hospital multi drug resistant bacteria infection monitoring and control efforts, and gradually reduce the incidence of multiple drug-resistant bacteria infection in hospital.2.Try to shorten the ICU length of stay, as soon as possible to pull out the ventilator, reduce the combined application of antibacterial drugs on patients with,should use medicine under the guidance of drug sensitivity test, on the basis of the disease more, low protein, fever patients to focus on monitoring, if there is a need for timely treatment of abnormal.3.Logistic regression model fit and predictive ability is better. |
PDF Full Text Request