Use of the API 20E, Oxi/Ferm, and Minitek systems to identify nonfermentative and oxidase-positive fermentative bacteria: seven years of experience.

The API 20E (API) system, the Oxi/Ferm (OXF) system, and the Minitek (MIN) system were used over a seven-year period to identify clinical isolates of nonfermentative bacteria (NFB) and oxidase-positive fermentative bacteria (OPFB). A total of 742 NFB and OPFB were tested with the API system, 988 with the OXF system, and 918 with the MIN system. The organisms represented 34 recognized species, nine alpha-numeric designates, and three species of unnamed NFB. Results were compared to those obtained with conventional test methods, which were considered correct in all instances. The accuracy of identification of NFB and OPFB was 61% with API, 80% with OXF, and 72% with MIN. Identification was more favorable for all systems with the oxidase-negative bacteria than with the oxidase-positive bacteria. API successfully identified 84% of the oxidase-negative organisms compared to 48% of the oxidase-positive organisms, OXF identified 93% compared to 75%, and MIN identified 96% compared to 61%. Failure of identification was usually a result of failure of the generated codes to appear in the compendia, rather than of misidentification, especially for the oxidase-positive organisms. The OXF system required the greatest number of supplemental tests for identification with a ratio of .65 (supplemental tests per isolate) and the API the least (.20). The API system generated the most codes with a ratio of .46 and the OXF system the fewest codes (.16).