ABSTRACT
The automated microdilution Sensititre System (Sensititre, LtD.) was evaluated for the identification of 120 clinically isolated fermenter and non-fermenter Gram negative bacilli and of 12 ATCC reference strains (American Type Culture Collection). Daily and overnight (5 and 18 hours) identifications were performed according to the manufacturer. In both cases, the results were compared with those obtained with the overnight API 20E (API System, La Balme les Grottes, Montalieu Vercieu) used as reference method. Concordance between the results obtained after 5 hrs of incubation and those obtained with API 20E was found in 65% of cases (acceptable identification criteria with probability over 80%). When results were determined after 18 hrs of incubation (identification criteria with probably over 80%) concordance was found in 82.5%. Reproducibility obtained by repeated (10 times) identification with four different strains was 100% both at 5 hrs and 18 hrs. Identification accuracy of the 12 ATCC strains was 8/12 (66.6%) after 5 hrs and 12/12 (100%) after 18 hrs.
Subject(s)
Gram-Negative Bacteria/isolation & purification , Evaluation Studies as Topic , Predictive Value of Tests , Reproducibility of ResultsABSTRACT
During one year period our laboratory carried out 859 hemocultures. These have been evaluated with two methods: a conventional biphasic method (Castaneda bottles), and the automated radiometric method (Bactec System). 185 cultures were obtained with one or both methods. Of these 3.5% were considered contaminated, therefore the clinically significant isolation rate was 16.2%. Of these 86.4% was recovered by biphasic system and 79.8% by Bactec System. The recovery time to positivity and spectrum of isolates were similar for the two methods. Although there were substantially more contaminants isolated in the Vacutainer-Bactec System.
