Search

An animal model of disseminated Mycobacterium haemophilum infection was utilized to compare treatment with azithromycin, ciprofloxacin, rifabutin, and the combination of clarithromycin with rifabutin. Following subcutaneous challenge with M. haemophilum, local and disseminated infection occurred only in immunosuppressed mice. For disseminated infection, ciprofloxacin was relatively ineffective therapy. Clarithromycin and rifabutin alone significantly reduced the tissue burden in the spleen after 4 weeks of therapy. Combination therapy with rifabutin and clarithromycin was superior to 4 weeks of treatment with the individual agents. When immunosuppressed mice were treated for 20 weeks with the combination of rifabutin and clarithromycin, the tissue burden remained reduced in the spleen at 1 month following the completion of therapy. Combined rifabutin and clarithromycin provide effective treatment for M. haemophilum in this model.

The activity of three aminoglycosides and six beta-lactam antibiotics on strains of Escherichia coli, Klebsiella pneumoniae, Proteus mirabilis, Pseudomonas aeruginosa, Staphylococcus aureus, and enterococci was studied. The minimal inhibitory concentrations (MICs), the minimal bactericidal concentrations (MBCs), and the minimal antibiotic concentrations (MACs) were determined after 5 h of incubation in broth cultures by colony-forming-unit counts. The MICs were also determined by agar dilution after 24 h of incubation. The MICs on agar after 24 h of incubation were higher than those in broth after 5 h of incubation. The differences ranged from 1.1- to 14.2-fold, but in most cases were only three- to fivefold (P less than 0.05 to less than 0.001). The MBCs at 5 and 24 h were comparable in 71% of tests. For current practice, the MBC of enterococci can be determined after 5 h of incubation with antibiotics. The aminoglycosides showed MBCs which were closer to the MICs than were those of the beta-lactam antibiotics, which required a higher multiple of the MIC to show a bactericidal effect. The MBCs of oxacillin and cefamandole for S. aureus after 5 h of incubation were greater than 128 times the respective MICs. The MACs ranged from 1/1.5 to 1/7 of the 5-h MICs. The three endpoints, MIC, MBC, and MAC, indicate the antibacterial range of an antibiotic in terms of inhibition of growth and bacterial survival. The data suggest that the antibacterial range of an antibiotic is similar for most strains of a given species and is, to some extent, a characteristic of similar antibiotics.

Exposure of strains of Escherichia coli to ampicillin and mezlocillin and of strains of Pseudomonas aeruginosa to azlocillin and cefsulodin caused the bacilli to elongate into filaments. The bacilli and their filaments were incubated with fresh human polymorphonuclear neutrophils (PMNs), and the phagocytic process was recorded by means of phase-contrast microscopy. The bactericidal effect of PMNs on both filaments and bacilli was quantitated by counts of colony-forming units. A single PMN phagocytized one or more filaments, some of which were as long as 90 micron and contained as many as 20 genomes. Two predominant patterns of phagocytosis of filaments were observed. When the ratio of bacteria to PMNs was low (0.2-1.8), the rate of killing was 62%-81%. When the ratio was higher (5-12), the rate of killing of both filaments and bacilli was lower. As an alternative to colony-forming units, cell mass was used as a gauge of phagocytic activity. The relative mass of killed filaments was considerably greater than that of killed bacilli; this finding indicated that filaments were much more susceptible than were bacilli to the bactericidal activity of PMNs.

Subinhibitory concentrations of antibiotics can produce in vitro aberrant forms of bacteria that are similar to those observed in specimens and cultures from patients being treated with antibacterial agents. Eight species of bacteria were grown on membranes placed on agar containing subinhibitory concentrations of nine antibiotics. The resulting organisms were examined by Gram stain and electron microscopy. Gram stains showed filamentous and granular forms of enterobacteria with bipolar staining, giant staphylococci, and rodlike pneumococci. Electron micrographs showed changes in the number and distribution of ribosomes in enterobacteria and septum abnormalities in cocci. Such abnormal forms can occasionally simulate the appearance of quite different species, and they may indicate the presence of a subinhibitory antibiotic concentration at the site of infection as a result of prior antibacterial therapy.

Five strains of Salmonella were grown on agar containing subinhibitory concentrations of ampicillin, gentamicin, tetracycline, or chloramphenicol, and on drug-free agar (control). Antigens were prepared by heating bacteria suspensions in saline solution at 100 C or by suspension in alcohol. Antigens were tested for agglutinability with somatic O antisera and fluorescent antibody staining. The alcohol-treated antigens prepared from organisms grown in the presence of antibiotics other than ampicillin had lower agglutination titers than did the control grown on drug-free agar. Heat-treated control antigens had lower agglutination titers than did the alcohol-treated control antigens. The agglutinability of heat-treated antigens prepared from organisms that had been exposed to antibiotics was preserved or enhanced. The agglutinated filaments produced a flocculent sediment, in contrast to the granular sediment of the controls. No significant morphologic abnormality in cell-wall structures of antibiotic-exposed cells could be detected by either electron microscopy or the fluorescent antibody staining.

Bacterial susceptibility to 16 commonly used antibiotics was analyzed for a 12-year period (from 1971 to 1982, inclusive). Susceptibilities of 5,828,243 strains isolated from a mean of 242 hospitals nationwide and of 194,575 strains isolated at the Massachusetts General Hospital, Boston, Mass., and the Bronx Lebanon Hospital Center, New York, N.Y., were compared. Strains of Escherichia coli, Staphylococcus aureus, Haemophilus influenzae, and Pseudomonas aeruginosa showed virtually the same susceptibilities to antibiotics throughout the 12-year period, whereas Streptococcus faecalis and Staphylococcus epidermidis showed significant increases in resistance to most antibiotics. The close similarity between antibiotic susceptibilities shown at both the 242 hospitals and the 2 individual hospitals suggests that this analysis accurately reflects trends of bacterial resistance to antibiotics in U.S. hospitals. Since most of the species analyzed produce serious disease and high mortality, their susceptibility to antibiotics is relevant both to physicians treating infectious diseases and to epidemiologists.

Biological specimens obtained from patients with infections that are treated with antibiotics at dosages that resulted in body fluid concentrations below those that are minimally inhibitory, exhibit abnormal forms of bacteria. Identical morphological changes can be induced in vitro when the parental normal forms of the bacteria. Identical morphological changes can be induced in vitro when the parental normal forms of the bacteria are exposed to subminimal concentrations of certain antibiotics. The phenomenon of induced abnormal morphology and ultrastructure has received consideration in terms of: (1) the relationship of the induced alterations of the specific bacterial species to the given antibiotic, (2) in vitro conditions required for antibiotic induced alterations, (3) the mechanisms of action; (4) immune responses to the abnormal forms, and (5) relationship between abnormal structure and the inhibition of bacterial replication. This review provides the reader with a description and the significance of current research in the above areas which together comprise the growing field of subinhibitory concentrations of antibiotics.

The thymidine incorporation method was compared with the disc diffusion and broth dilution methods for determining the antibiotic susceptibility of Escherichia coli (233 strains), Proteus mirabilis (72 strains) and Klebsiella species (189 strains). The method measures changes in the rate of DNA synthesis of bacterial strains exposed to an antibiotic by comparing the amount of radioactive thymidine incorporated by these strains in relation to control strains. Good agreement was obtained between the results of the thymidine incorporation and disc diffusion methods for all antibiotics tested except streptomycin and tetracycline. Minimal inhibitory concentrations of antibiotics determined by the thymidine incorporation method after 3 and 5 hours were generally lower than those obtained after 24 hours by the conventional broth dilution procedure. The thymidine incorporation method may be potentially useful for determining the antibiotic susceptibility of bacteria a few hours after inoculation of a clinical specimen.

Cultures of Staphylococcus aureus, Neisseria gonorrhoeae, and Streptococcus bovis were incubated on membrane filters on agar containing antibiotics at one-third to one-fourth the minimal inhibitory concentration for the particular bacterial strain. S. aureus was grown in the presence of oxacillin, cephaloridine, or lincomycin. N. gonorrhoeae and S. bovis were grown in the presence of penicillin. The membranes were then incubated in drug-free agar, after which viability was determined and the cells were examined by electron microscopy. S. aureus exposed to oxacillin and cephaloridine grew into cells two to seven times larger than normal that contained thick multiple cross walls. S. aureus exposed to lincomycin grew into cells 1.5 to 2 times larger than normal, with multiple thick cross walls and periheral cell walls twice the normal thickness. N. gonorrhoeae cells exposed to penicillin were slightly larger than normal and had cross walls that were up to eight times thicker than normal. After transfer to drug-free agar, cells became smaller, and some normal organisms could be seen. S. bovis incubated in the presence of penicillin grew into filaments that contained no cross walls. Two hours after the return to drug-free agar, filaments with cross walls as well as normal cells were observed. Exposure to subinhibitory concentrations of penicillin did not affect the growth of the peripheral cell wall of S. aureus, N. gonorrhoeae, or S. bovis, but appeared to inhibit lysis of cross walls in S. aureus and N. gonorrhoeae and to inhibit the synthesis of cross walls in S. bovis; that is, the rates of peripheral and cross wall formation differed in their susceptibility to penicillin. These facts suggest that peripheral growth and cross wall formation in cocci are separable processes.