Epidemiological Studies Of C. Difficile Multilocus Sequence Typing And Risk Factors

Objective:1. To isolate Clostridium difficile(C. difficile) from hospitalized ICU patients with antibiotic-associated diarrhea (AAD), and to study the features of C. difficile toxin. Using multilocus sequence analysis (MLST), random primer PCR analysis (RAPD) to analysis the C. difficile genotype, the predominant strain, and to compare if RAPD method is suitable for typing C. difficile with MLST as a reference. By MLST is the international typing method, the research data can be compared with international strains for MLST database.2. To analyse C. difficile isolating rate from AAD patients, incidence of CDAD in hospitalized patients in the ICU, compare and analyse the differences in AAD and CDAD in patients with risk factors by multivariate analysis. For clinical treatment, prevention, monitoring CDAD, and to develop prevention and treatment programs to provide reasonable data support.Methods:Stool specimens from patients with AAD in comprehensive ICU, neurology ICU, neurosurgery ICU, respiratory ICU were cultured for C. difficile by using CDMN medium during June2013and February2014. Identification of C. difficile by conventional methods and molecular biology techniques. Then the toxin A, B gene tcdA and tcdB, binary toxin genes cdtA and cdtB were amplified. Seven housekeeping genes of C. difficile were amplified and sequenced genes obtained by MLST. Using RAPD method for C. difficile typing, cross checker drawn matrix, NTsys2.10e software clustering analysis based on the function tree like figure. MLST and RAPD data and basic informations on C. difficile were edited by using bionumerics software. Patient-days of stay in ICU for patients as the denominator, to caculate the incidence of hospital-onset/hospital-associated CDAD (HOHA CDAD).By filling out the questionnaire and using Epidate entry, data were analyzed by using SPSS13.0for single-factor measurement data. Using t-test, chi-square test count data. Multi-factors were analyzed by using Backward:LR logistic regression analysis.Results:(1)40C. difficile were isolated from725specimens for394patients in4ICU during9months. Mainly from the Comprehensive ICU (n=12), Neurology ICU(n=230, Neurosurgery ICU(n=50), C. difficile was not detected in Respiratory ICU.25strains were detected for tcdA,33strains were detected for tcdB.25strains were tcdA+tcdB+C. difficile, accounting for62.50%,8strains were tcdA-tcdB+, accounting for20.00%,7strains were tcdA-tcdB-, accounting for17.50%. One strain was detected cdtA and cdtB gene.(2) A total of9STs were generated from40strains, belonged to three clade. ST54(n=9), ST39(n=7), ST26(n=6), ST3(n=4) were the dominant sequence type. ST201were detected positive for the binary toxin, and ST274was the new type.There are27types analysed by RAPD genetic fingerprint classification, the most common type was R20(n=5), followed by R13(n=4), R6(n=4). The RAPD fingerprints were clustered analysis with ST found that the R20corresponding with ST26, R13corresponding with ST54, and R6VS ST37, due to differences in individual band, ST39distributed between R1-R5, and R1-R5mutual relationship were the subtypes.(3) The incidence of HOHA CDAD was16.60/10,000patient-days, and it was42.82/10,000patient-days in Neurology, CDAD patients accounted for 8.12%(32/394) of AAD.(4) CDAD was contrasted with AAD by univariate analysis, there were4risk factors, prescribed with antibiotics in the past one month (x2=12.30, P=0.00), parenteral nutrition (x2=4.72, P=0.03), hospitalization days before diarrhea (t=-2.12, P=0.03), antibiotics use days (t=-1.99, P=0.05). By ultivariate logistic regression analysis, prescribed with antibiotics in the past one month, long hospitalization days before diarrhea were the risk factors for CDAD. The odds ratio of prescribed with antibiotics in the past one month was7.30, The odds ratio of hospitalization days was2.28by rising an additional level.Conclusion:(1) The toxin of A+B+strains were the main types, A+B-were not uncommon also.(2) The main MLST types were ST54, ST39, ST26.(3) The molecular epidemiological analysis of RAPD for C. difficile was high resolution.(4) Incidence of HOHA CDAD was high for the ICU patients.(5) CDAD compared with the AAD has significant risk factors.