ABSTRACT
Shortening the turnaround time of antimicrobial susceptibility testing (AST) of bacteria permits a significant reduction of patient morbidity, mortality, and cost. Conventional blood culture methods are the gold standard diagnostic test to guide management of patient with sepsis, but the conventional process requires at least 12 to 24 h after the blood culture has been flagged as positive due to requirement for pure colonies. We describe a simple and inexpensive method to obtain faster AST with MicroScan system (Beckman Coulter) directly from positive blood cultures. Conventional and direct identification and AST were performed simultaneously by both methods in 1070 blood cultures, and 9106 MICs were determinated. About 96.5% were correctly identified with the direct method. Overall, categorical agreement was 92.86%. We found 46 very major errors, but globally the results showed a good correlation with the standard method, particularly favorable for E. coli and K. pneumoniae, except amoxicillin-clavulanate and piperacillin-tazobactam. For P. mirabilis, betalactams antibiotics (except second- and third-generation cephalosporines) showed a good correlation, and also a good correlation was found for ciprofloxacine and gentamicine in P. aeruginosa and amoxicillin-clavulanate, ciprofloxacine, gentamicine, and cotrimoxazole in E. cloacae. This method has the main advantage of providing reliable results 1 day earlier, being a simple, fast, and cheap method for identification and antimicrobial susceptibility testing results from positive blood cultures.
Subject(s)
Anti-Bacterial Agents/pharmacology , Gram-Negative Bacterial Infections/microbiology , Anti-Bacterial Agents/therapeutic use , Blood Culture , Escherichia coli/drug effects , Gram-Negative Bacterial Infections/drug therapy , Humans , Klebsiella pneumoniae/drug effects , Microbial Sensitivity Tests , Point-of-Care Testing , Reproducibility of ResultsABSTRACT
Limited information exists on the performance of antigen-based rapid influenza diagnostic tests (RIDT) in diagnosing the novel influenza A pandemic (H1N1) 2009 virus. Large studies evaluating these tests in consecutive patients with a broad clinical spectrum of influenza-like illnesses are needed. We assessed the ClearView® Exact Influenza A & B test (Inverness Medical, Cologne, Germany) in comparison with real-time (r)RT-PCR for detection of the novel influenza A (H1N1) in a population-based prospective study of 1016 adults and children with suspected influenza in Spain. Three hundred and one (29.6%) patients had a positive sample with the rRT-PCR assay for influenza A and B viruses, with 297 (29.2%) confirmed cases of the novel influenza A pandemic (H1N1) 2009 virus. Fifty (16.8%) patients with confirmed A (H1N1) 2009 virus were admitted to hospital, with six of them to the intensive care unit. In comparison with rRT-PCR, the ClearView® Exact Influenza A & B test had a sensitivity of 19% (95% CI 14-23), a specificity of 100%, a positive predictive value of 100%, and a negative predictive value of 75% (95% CI 72-78). The sensitivity of the test remained low across all demographic and clinical strata. Although a positive RIDT performed well in predicting PCR-confirmed infection with pandemic H1N1 virus, the sensitivity was very low and a negative test result was a poor predictor of the absence of infection.
