RESUMO
Objective To explore specimen sampling for microbial culture in ICU patients with documented infections in order to offer clinical evidence for improving the rational use of antibiotics.Methods Patients with documented infection on the first day after admission into ICU and discharged from ICU from July to December 2012 and from July to December 2013 were enrolled in the study.Clinical data including presence or absence of infection,initial antimicrobial therapy,microorganism specimen sampling and culture were retrospectively analyzed.Results Of 841 patients discharged from ICU,443 had evidence of infections and received antimicrobial therapy on the admission day,and only 30 (6.8％) of them had microbiological detection results prior to treatment.There were microbial specimens available at infection sites on the admission day in 369 cases,and 360 cases (97.6％) of them were sampled in the first three days after ICU admission,while only 119 cases (33.1％) were sampled before the first dose of antimicrobial therapy.Specimens sampled were sputum (56.4％) in the majority,followed by the blood (17.4％).Further analysis of 269 infected patients receiving initial broad-spectrum antimicrobial therapy also showed that only 33.5％ cases were sampled before the first dose of broad-spectrum antimicrobial administration.The positive isolation rate of multi-drug resistant isolates including A.baumannii,S.maltophilia and B.cepacia from specimens sampled after first dose of initial broad-spectrum antimicrobial therapy were significantly higher than those sampled before antimicrobial therapy,P ＜ 0.05.There was no significant difference in isolation rate of Staph.aureus and Enterobacteriaceae between samples obtained before and after first dose of initial broad-spectrum antimicrobial therapy.Conclusions Few evidence of pathogenic microorganisms was available before initial antimicrobial therapy in ICU patients.Although sampling rate of microbial specimens is high,the most of them are sampled after the first dose of antimicrobial administration,and the patentially contaminated specimens such as sputum in predominance,obviously decrease the reliability of authentic results obtained from microorganism culture.