Comparative evaluation of the API 20S system and the automicrobic system gram-positive identification card for species identification of streptococci

Two commercial methods, the API 20S system (API; Analytab Products, Inc., Plainview, N.Y.) and the Gram-Positive Identification Card (GPI; Vitek Systems, Inc., Hazelwood, Mo.), were evaluated without additional tests for the identification of 241 streptococcus strains. Organisms included 60 beta-hemolytic strains, 36 group D strains, 26 Streptococcus pneumoniae strains, and 119 viridans streptococcus strains. API correctly identified to species 68.3% of beta-hemolytic strains, 86.1% of group D strains, 53.9% of S. pneumoniae strains, and 12.6% of viridans streptococci. This method provided excellent identification of group A and B and S. faecalis strains. Overall, API correctly identified 41.9% of strains to species, with 41.1% good likelihood but low selectivity, 15.8% incorrect, and 1.2% not identified. GPI correctly identified to species 58.3% of beta-hemolytic strains, 97.2% of group D strains, 80.8% of S. pneumoniae strains, and 57.2% of viridans streptococci. Group A, B, and D strains were all accurately identified by this system. Overall, GPI correctly identified to species 66.0% of strains, with 8.7% correct preliminary identification, 20.8% incorrect, and 4.6% not identified. Both methods represent a worthwhile advance in streptococcal identification. Neither system, however, can be recommended for species identification of the viridans group at this time.