CURRENT STATUS OF COMBINED ANTIBIOTIC THERAPY

Abstract

The use of combinations of drugs in "shotgun" fashion, on the theory if one will do something, two or three will accomplish more, is to be deplored. A single antibiotic can be used effectively in most infections caused by a single organism. Furthermore, a single wide-spectrum antibiotic may be used in many mixed infections. In certain infections, the value of combinations of antibiotics has been proved, both in the laboratory and in the clinic: (a) streptomycin plus one of the tetracyclines in brucellosis; (b) penicillin plus streptomycin in enterococcic endocarditis; (c) erythromycin plus chloramphenicol in serious staphylococcal infections in which the organism is resistant to penicillin; (d) streptomycin, isoniazid and para-aminosalicyclic acid in treatment of tuberculosis. In these infections, the proper combination should be used from the start and in full therapeutic dosage. Mixtures of antibiotics may occasionally be useful in individual cases outside this group but, in general, these mixtures do not produce a synergistic effect. If the infection does not fall into one of the four classes already cited, the in-vitro effect of combinations of various antibiotics should be studied, providing the patient's illness is such that a delay of 48 to 72 hours is warranted. The combination showing the greatest synergistic effect should then be used. Persisting urinary tract infections and endocarditis are examples of conditions in which this method is likely to produce results. Even here, it must be borne in mind that such in-vitro tests do not guarantee that a certain mixture of antibiotics will be effective clinically; in fact, they may be misleading. In accord with the recommendations of Dowling, if there is not sufficient time for an in-vitro study to be carried out, two antibiotics in Jawetz and Gunnison's Group 1 (see text) may be used together, if each alone is partially effective against the causative organism. If no two antibiotics in Group 1 fit this criterion, and a combination of an antibiotic from Group 1 and one from Group 2 does fit the criterion, this combination should be given in doses that will result in full therapeutic concentrations of each antibiotic at the site of infection. For delaying the emergence of resistant strains of tubercle bacilli, combinations of two or more of the following drugs are indicated: Streptomycin, isoniazid and para-aminosalicylic acid. A combination of chloramphenicol with erythromycin is also indicated to delay emergence of resistance of staphylococci to the batter antibiotics. In the treatment of seriously ill patients before a bacteriologic diagnosis is available, two or more antibiotics may properly be administered. Such illnesses include endocarditis, suspected staphylococcal pneumonia in infants, tuberculosis, brucellosis, and meningitis due to an unidentified organism. Combined antimicrobials should be given only after a careful clinical diagnosis has been made, and in doses that would be optimal for each drug if used alone. Readymade mixtures are not recommended for use systemically; certain combinations of agents, such as a mixture of bacitracin and polymyxin B, may have a place in topical therapy. Finally, there are several potentially harmful or undesirable effects that may result from the use of combinations of antimicrobials and these must be considered whenever their possible use arises. These include: (1) the tendency of fixed, "packaged" combinations to encourage inadequate therapy; (2) the possible increase in hypersensitivity and/or toxicity to one or more of the agents in a combination; (3) the probable emergence of bacterial resistance to either or both of the antibiotics in a mixture; (4) superinfection by originally resistant organisms not affected by the therapy; (5) the accumulation of antibiotic-resistant organisms within hospitals or other semiclosed communities; and (6) the possibility of interference of one antibiotic with the operation of another in a given combination.