Cefdaloxime (formerly RU29246; Hoechst-Roussel Pharmaceuticals Inc., Somerville, N.J.) a new active component of the HR-916 ester, was tested by dilution and two disk (10- and 30-mu-g) diffusion susceptibility tests against 391 clinical isolates. Interpretive criteria were proposed for three potential MIC breakpoints of less-than-or-equal-to 1, less-than-or-equal-to 2, and less-than-or-equal-to 4-mu-g/ml. Analyses by regression line and error rate bounding methods minimized false-susceptible (very major) errors and produced a greater-than-or-equal-to 90% absolute interpretive agreement between susceptibility test methods. The less-than-or-equal-to 2-mu-g/ml breakpoint seemed optimal when 10-mu-g disks and the available human pharmacokinetics were used. The following inhibition zone diameter criteria were proposed: susceptible, greater-than-or-equal-to 19 mm; resistant, less-than-or-equal-to 15 mm. These recommendations for clinical trials should remain tentative until additional information about cefdaloxime formulations, pharmacokinetics, and patient outcomes can be correlated with in vitro susceptibility test results.