OBJECTIVE: To compare the efficacy of four phenotypic methods for the identification of strains producing extended-spectrum β-lactamases isolated from urine cultures. MATERIALS AND METHODS: Comparative cross-sectional study. 147 strains isolated from positive urine cultures for Escherichia coli, Klebsiella pneumoniae and Proteus mirabilis between January and February 2009 in the National Institute of Health of Children underwent a screening test, those which resulted positive were processed for confirmatory testing through the four phenotypic methods evaluated. RESULTS: Out of the 147 strains, 43 (29.3%) were suspicious in the screening tests. Using the method described by the Clinical and Laboratory Standards Institute (CLSI) as a reference standard for this study, 27 strains (62.8%) were positive, with similar results using Jarlier's method. On the other hand, using Hodge's and tridimensional methods 23 (53.5%) of samples were positive. The evaluation of the confirmation methods in comparison to the one described by CLSI, showed a sensitivity and specificity of 100% for Jarlier's method, on the other hand, for Hodge's and tridimensional methods we found a sensitivity of 85.2% and 100%, respectively. CONCLUSIONS: All the evaluated methods showed a high efficacy, without significant differences, so they could all be used according to the available facilities of each clinical laboratory. Nevertheless, due to its advantages besides the technical aspect, like costs, easiness and feasibility of its application, we recommend the use of Jarlier's method.
OBJECTIVE: To compare the efficacy of four phenotypic methods for the identification of strains producing extended-spectrum β-lactamases isolated from urine cultures. MATERIALS AND METHODS: Comparative cross-sectional study. 147 strains isolated from positive urine cultures for Escherichia coli, Klebsiella pneumoniae and Proteus mirabilis between January and February 2009 in the National Institute of Health of Children underwent a screening test, those which resulted positive were processed for confirmatory testing through the four phenotypic methods evaluated. RESULTS: Out of the 147 strains, 43 (29.3%) were suspicious in the screening tests. Using the method described by the Clinical and Laboratory Standards Institute (CLSI) as a reference standard for this study, 27 strains (62.8%) were positive, with similar results using Jarlier's method. On the other hand, using Hodge's and tridimensional methods 23 (53.5%) of samples were positive. The evaluation of the confirmation methods in comparison to the one described by CLSI, showed a sensitivity and specificity of 100% for Jarlier's method, on the other hand, for Hodge's and tridimensional methods we found a sensitivity of 85.2% and 100%, respectively. CONCLUSIONS: All the evaluated methods showed a high efficacy, without significant differences, so they could all be used according to the available facilities of each clinical laboratory. Nevertheless, due to its advantages besides the technical aspect, like costs, easiness and feasibility of its application, we recommend the use of Jarlier's method.
Authors: Anika Larson; Stella Maria Hartinger; Maribel Riveros; Gabriela Salmon-Mulanovich; Jan Hattendorf; Hector Verastegui; Maria Luisa Huaylinos; Daniel Mäusezahl Journal: Am J Trop Med Hyg Date: 2019-06 Impact factor: 2.345