Ticarcillin and ticarcillin-clavulanic acid susceptibility tests: error rates for disk tests with consecutively isolated members of the family Enterobacteriaceae.

A five-laboratory coordinated study was undertaken to determine whether ticarcillin and ticarcillin-clavulanic acid disk susceptibility tests could accurately detect resistance among enteric bacilli. Each facility performed disk tests and broth microdilution susceptibility tests with reagents distributed from a common source, and appropriate controls were included in order to ensure methodologic uniformity. Each institution tested 500 consecutively isolated enteric bacilli against ticarcillin and ticarcillin-clavulanic acid. The prevalence of discrepancies between disk tests and dilution tests with 2,435 unselected enteric bacilli provided a valid estimate of the true error rate for tests with the two disks. The current interpretive criteria for susceptibility tests with ticarcillin and ticarcillin-clavulanic acid disks were found to be reliable; i.e., there were major or very major discrepancies between MIC categories and disk test results for less than 1% of all strains and minor discrepancies for only 5 to 10%. Even lower error rates would occur if the zone size-interpretive criteria were modified, but at this time it is difficult to determine whether the practical problems created by making such changes can be justified by the magnitude of the problem that is being resolved.