Study On Association Of Carbapenem-resistant Organism Intestinal Colonization And Host Lymphocyte Function With Clinical Infection

?Objective?Carbapenem-resistant organisms(CRO)have attracted much attention due to the considerable clinical infection rate and mortality.Early detection of carriers and infected patients and take appropriate infection prevention and control measures are import to CRO control.Carbapenem-resistant Enterobacteriaceae(CRE)is an important CRO.CRE colonization and host impaired immune function are independent risk factors for infection and death.In order to provide theoretical basis and guidance for the prevention measures against subsequent infection and reduce the incidence of clinical infection and mortality,we aim to determine the homology of CRE rectal colonization strains and infection strains,track and trace the occurrence path of CRE parenteral infection,and analyse the association of colonization strains with subsequent infection and the changes of host lymphocyte function in the development of infection.?Methods??.Study on homology between CRE colonization strains and CRE infection strainsRectal swabs were obtained from high-risk patients hospitalized in intensive care unit(ICU)and hematology department.Using Klebsiella pneumoniae carbapenemase(KPC)-screening agar to screen CRE colonization and monitoring the infection of patients.Patients combined with CRE colonization and CRE infection were recruited.The colonization strains and infection strains were collected,as well as patients'clinical information.Bacterial species were identified using matrix-assisted laser-desorption ionization time-of-flight mass spectrometry(MALDI-TOF MS).Antibiotic susceptibility testing was performed by the broth microdilution.Carbapenemase genes and MLST type were identified by polymerase chain reaction(PCR).Pulsed field gel electrophoresis(PFGE)and whole genome sequencing(WGS)were performed to analyse the genetic relationship between colonization strains and infection strains.Combined with the clinical information of patients,the relationship between CRE colonization and infection was analysed.?.Risk assessment and prediction model of CRE infection in CRE colonized patientsPatients with CRE colonization in ICU were recruited.Collecting their clinical information and microbiological factors of the colonized strains,including Charlson Comorbidity Index(CCI),invasive procedures,use of immunosuppressant,antibiotic treatment and species,antibiotic resistance phenotype,carbapenemase type of colonized strains,etc.Patients were grouped according to whether occur subsequent CRE infection during hospitalization and matched by propensity score matching(PSM).Related risk factors were subjected to univariate analyses.Infection risk predictive models were established based on binary logistic regression analyses,to assess the risk of secondary CRE infection following CRE colonization in ICU inpatients.?.Study on association of host lymphocyte function with CRE colonization and infectionPatients with negative CRE-screening,CRE colonization,and CRE infection were recruited.Once CRE screening is diagnosed with negative/positive,or CRE is detected in infected patients,peripheral blood of patients was collected.The number and IFN-?producing ability of lymphocytes were detected by flow cytometer.Collecting clinical information of patients,including prior hospitalization,CCI score,use of immunosuppressant,invasive procedures,antibiotic exposure,etc.The patients were matched by PSM.The number and IFN-?producing ability of lymphocytes between the groups were analysed,as well as the relationship of clinical factors between CRE colonization and infection.?.Risk assessment and prediction model of adverse prognosis in CRO infected patientsCRO-infected patients were recruited and their peripheral blood was collected as soon as infection was confirmed.The number and IFN-?producing ability of lymphocytes were detected by flow cytometer.Related clinical,routine laboratory and microbiological risk factors of patients were investigated,including APACHE II score,use of immunosuppressant,invasive procedures,antibiotic treatment,blood routine,liver and kidney function indicators,CRE colonization,infection type,co-infection,etc.Patients were grouped according to the 30-day mortality.Related risk factors were subjected to univariate analyses.Mortality risk predictive models were established based on binary logistic regression analyses,to identify high-risk infected individuals in the early stage of CRO infection.?Results??.Study on homology between CRE colonization strains and CRE infection strainsFrom May 2018 to January 2019,a total of 31 patients combined with CRE colonization and CRE infection in ICU and hematology department.The colonization strains and infection strains were 62 and 33,respectively.According to the bacteria species,carbapenemase gene,ST type and antimicrobial resistance phenotype of 31 cases,10 cases were selected for PFGE(42 isolates)and WGS(32 isolates)analysis.The 42 CRKP were divided into 11 clusters by PFGE.The 32 CRKP were divided into 2 branches by cluster analysis of drug resistance gene and 5 clusters by the SNP phylogenetic tree.Combined with the clinical information of the patients,the colonization strains of Case 1 were highly homologous to the infected strains,but there were differences in antimicrobial resistance phenotypes and drug resistance gene.There were differences among the colonization strains of Case 2,Case 3 and Case 4.Case 5 and Case 10 were subsequent infection following CRE colonization,because the colonized and infected strains were highly homologous.Case 8 and Case 9 were also highly homologous,but the colonized strains were isolated after infection.The colonization strains of Case 6 and Case 7 were far related to their infected strains,and there were two isolates of CRKP in the clinical infection specimen of Case 6.?.Risk assessment and prediction model of CRE infection in CRE colonized patientsFrom May 2018 to August 2020,the CRE colonization rate of ICU inpatients was 8.9%,and the infection rate of colonized patients was 11.0%.After 3:1 case matching,a total of 108 CRE colonization patients were included for analysis,among which 81 patients without subsequent CRE infection were included in the control group and 27 patients with infection were included in the infected group.The proportion of colonization strains with bla _KPC,MIC_IPM?32?g/ml and MIC_ETP?64?g/ml in the infected group was higher than that in the control group.After multivariate analysis combined with clinical and microbial factors,mechanical ventilation(adjusted odds ratio[a OR],5.044;95%confidence interval[CI],1.114 to 22.832;P=0.036),surgery(a OR,5.764;95%CI,1.718 to 19.332;P=0.005),?-lactamase inhibitor(a OR,0.141;95%CI,0.042 to 0.465;P=0.001)and bla _KPC(a OR,5.133;95%CI,1.598 to 16.493;P=0.006)was independently associated with subsquent infection following CRE colonization.AUC analysis showed that the combined model was superior to the model including clinical factors or microbial parameters separately.When the cut-off value of was 0.501,the accuracy,sensitivity and specificity of the combined model were 86.11%,63.0%and 93.8%,respectively.?.Study on association of host lymphocyte function with CRE colonization and infectionFrom November 2018 to August 2020,60 patients with negative CRE-screening,CRE colonization,and CRE infection were included uniformly after case matching.There was no statistical difference in lymphocyte number between negative control group and CRE colonization group or between CRE colonization group and CRE infection group.The IFN-?producing ability of NK cells in CRE colonization group was significantly lower than that in negative control group(P<0.001),and he IFN-?producing ability of CD4⁺T cells and CD8⁺T cells in CRE infected group was lower than that in CRE colonization group(P=0.002,P=0.040).There was statistical difference in the use of immunosuppressant,white blood cell count,neutrophils ratio and antibiotic exposure between negative control group and CRE colonization group.In addition to above factors,there also was statistical difference in prior hospital,transfer unit,gastrointestinal injury,dialysis and various invasive procedures between CRE colonization group and CRE infection group.?.Risk assessment and prediction model of adverse prognosis in CRO infected patientsFrom November 2018 to August 2019,a total of 127 CRO-infected patients were enrolled,including 85 survivors and 42 non-survivors.The number and IFN-?producing ability of lymphocytes were remarkably decreased in non-survivors.The number of IFN-?⁺CD4⁺T cells could effectively predict 30-day mortality of CRO infection.Its area under the receiver operating characteristic(ROC)curve,sensitivity,specificity and accuracy,were 0.889(95%CI,0.834-0.945),81.0%,80.0%and 80.3%,respectively.In multivariate analysis of laboratory parameters,IFN-?⁺CD4⁺T cell number and creatinine concentration were selected for the2-marker model to predict prognosis fleetly.Its area under the ROC curve,sensitivity,specificity and accuracy were 0.894(95%CI,0.841-0.947),83.3%,82.4%and 82.7%,respectively.?Conclusions?1.There may be polyclonal colonization of CRE in the intestines of CRE colonizers,and the colonization is dynamic.The relationship between CRE colonization and parenteral infection is complex.The infected strains may be the colonization strains,or may be unrelated to the colonization strains,and CRE infection can be converted to CRE intestinal colonization.2.If the colonization strain was with bla _KPC,MIC_IPM?32?g/ml or MIC_ETP?64?g/ml,patients with CRE colonization are at higher risk for subsequent CRE infection.A predictive model combining clinical factors and microbial parameters was successfully constructed,which could effectively predict the risk of subsequent infection following CRE colonization in ICU inpatients.3.The occurrence of CRE colonization may be related to the impaired innate immune function of the patients,while the further impairment of acquired immune function may be related to the occurrence of CRE infection.In the development of CRE infection,the function of lymphocyte decreased in different degrees,and its changing trend and degree could reflect host immunity state and the progression of infection disease.4.Low lymphocyte count and impaired lymphocyte function are the key factors affecting the prognosis of CRO-infected patients.IFN-?⁺CD4⁺T cell number can effectively indicate 30-day mortality of CRO infection,as well as has potential value in monitoring prognosis.A 2-marker model was successfully established based on the combination of IFN-?⁺CD4⁺T cell number and creatinine,and it not only shows good performance in predicting prognosis of CRO-infected patients but also has potential value in monitoring prognosis during therapy.