Combined antibiotic therapy in serious infections caused by drug-resistant staphylococci (1956-1957)

Variation in performance of the serum bactericidal test

Questionnaires were sent to 37 diagnostic microbiology laboratories to assess their methods for performance of the serum bactericidal test. Of the 28 responding laboratories, 26 utilized the test, with considerable variation among methods. Culture and dilution media included Trypticase soy, Mueller-Hinton, brain heart infusion, Columbia broth, dextrose phosphate, and pooled human serum. Several different schemes for dilution of the test serum were employed, including both micro and macro methods. The greatest variation occurred in the size of the bacterial inoculum, with only eight laboratories having a method for standardization of the inoculum. Other differences included the time of incubation, the method of subculture, and determination of the bactericidal end point. These studies emphasize the need for assessment of the variables in the performance of this test and for the development of a standard method for diagnostic laboratories.

The "tough case" of bacterial endocarditis

Many cases of bacterial endocarditis can now be cured rapidly and easily. However, there remains a group of patients in whom the eradication of the cardiac infection presents great problems. Some principles of treating these "tough cases" of endocarditis are outlined by means of illustrative patients. Stress is placed on early diagnosis, isolation of the etiologic microorganism, determination of bactericidal antibiotics or combinations, and use of the serum bactericidal test to establish an adequate dosage regimen.

Serum bactericidal activity as a therapeutic guide in severely granulocytopenic patients with gram-negative septicemia

The peak and trough levels of bactericidal activity of the serum of 74 severely granulocytopenic patients (less than or equal to 500 polymorphonucleates per microliter) with hematologic malignancies and Gram-negative septicemia were measured using the patient's infectious organism and serum containing the given antibiotics. When the peak titer of bactericidal activity in the serum was greater than 1:8 the septicemia was cured in more than 90% of the cases. However, in order to achieve a satisfactory rate of cure, patients with less than 100 polymorphonucleates/microliter required higher peak levels than patients with 100-500 polymorphonucleates/microliter. Serum bactericidal activity was influenced by the in vitro susceptibility of the offending pathogen and by the presence of in vitro synergism between the given antibiotics. These two variables showed a correlation with the clinical outcome that proved to be increasing with the degree of granulocytopenia. Furthermore, synergistic combination of the antibiotics appeared essential when the in vitro susceptibility shown by the offending pathogen was moderate. These data suggest (i) that determination of the bactericidal activity of the serum may prove to be a useful method to predict the clinical outcome in severely granulocytopenic patients with Gram-negative septicemia; and (ii) under the same conditions, antibiotic combinations that have demonstrable in vitro synergy against the offending pathogen should be given the utmost consideration.

Medium-dependent variation in bactericidal activity of antibiotics against susceptible Staphylococcus aureus

Staphylococcus aureus resistant to bactericidal activity of antibiotics caused sepsis in three patients. Bacteriological and clinical responses were not achieved until serum and tissue fluid levels of administered antibiotics exceeded the minimum bactericidal concentration (MBC) of the infecting organism. Fifteen clinical isolates of S. aureus were tested in brain heart infusion broth and Mueller-Hinton broth for the MBC of gentamicin, vancomycin, clindamycin, oxacillin, cefazolin, and cephalothin. Results showed significant eightfold or greater broth-dependent differences in the MBC of at least one antibiotic against 87% (13/15) of strains tested. The MBC was unpredictable and varied with the strain, antibiotic, and medium used. No controlled studies are available to indicate the clinical significance of the MBC demonstrated in different media. The necessity for treating serious infection with bactericidal drugs has not yet been established; however, in septicemia such as that caused by bacterial endocarditis, bacteriostatic antibiotics have generally failed to eradicate the infection, whereas bactericidal agents have often been curative. Therefore, in patients unresponsive to usual antistaphylococcal therapy, we suggest that MBC testing be performed in at least two media and that treatment be instituted with antibiotics demonstrating the lowest MBC in all media used.