RESUMO
A total of 66 Acinetobacter isolates obtained from JIPMER hospital wards were subjected to phenotypic identification schemes involving 25-test and a simplified 13-test panel of carbon utilization or assimilation tests. Reference strains belonging to different DNA groups (n=24) were also tested. Identification was done using numerical approach based on a matrix constructed of phenotypic data published elsewhere and the strains were assigned to different DNA groups according to classification of Tjernberg & Ursing. Sixty-six strains tested represented 10 DNA groups in matrix of large test panel; at a probability level of 0.95. Much simplified scheme of 13 assimilation test panel failed to differentiate some isolates with in A. calcoaceticus-A. baumannii complex (Acb-complex) unlike extended panel. In all, from the large panel 95% of isolates were identified correctly among all the isolates and it did not identify 5% of isolates. From the small panel, a total of 89% of isolates were identified correctly and it could not identify 11% of isolates. Reduced number of assimilation tests to 13 from the large panel bought reduction in identification percentage rate by only 6%. It is impossible for many bacterial diagnostic labs worldwide to perform large panel of carbon utilization tests in routine practice. Simplified panel of assimilation tests suggested here seems to be the best alternative method for identification of Acinetobacter species.