The emergence of bacterial resistance and its influence on empiric therapy.

The discovery of antimicrobial agents had a major impact on the rate of survival from infections. However, the changing patterns of antimicrobial resistance caused a demand for new antibacterial agents. Within a few years of the introduction of penicillin, the majority of staphylococci were resistant to that drug. In the 1960s the production of the semisynthetic penicillins provided an answer to the problem of staphylococcal resistance. In the early 1960s most Escherichia coli were susceptible to the new beta-lactam antibiotic ampicillin; by the end of that decade, plasmid-mediated beta-lactamase resistance was found in 30%-50% of hospital-acquired E. coli. Use of certain agents resulted in the selection of bacteria, such as Klebsiella, that are intrinsically resistant to ampicillin. The original cephalosporins were stable to beta-lactamase, but the use of these agents was in part responsible for the appearance of infections due to Enterobacter species, Citrobacter species, and Pseudomonas aeruginosa. These bacteria, as well as Serratia, were resistant to many of the available beta-lactam agents. Aminoglycosides initially provided excellent activity against most of the facultative gram-negative bacteria. However, the widespread dissemination of the genes that cause production of the aminoglycoside-inactivating enzymes altered the use of those agents. Clearly, the evolution of bacterial resistance has altered the prescribing patterns for antimicrobial agents. Knowledge that beta-lactam resistance to ampicillin or cephalothin is prevalent is causing physicians to select as empiric therapy either a combination of two or more agents or agents to which resistance is uncommon. The new cephalosporins offer a broad spectrum of anti-bacterial activity coupled with low toxicity. However, physicians must closely follow the changing ecology of bacteria when these agents are used, because cephalosporins can also select bacteria resistant to themselves and thereby abolish their value as empiric therapy.