It is critical that new antibiotics targeted at multidrug-resistant pathogens be used judiciously in order to preserve their usefulness.

The discovery (and development) of antibiotics was undoubtedly one of the greatest advances of modern medicine. It is difficult for most of us to appreciate fully the impact that antibiotics have had in alleviating human suffering and prolonging life over the past half century. A discovery of comparable magnitude in current times would be that of the cure for cancer. Unfortunately, the emergence of antibiotic-resistant bacteria, particularly over the last decade, is now threatening the effectiveness of many antimicrobial agents. Nowhere is this threat greater than in hospitals where the sickest patients are congregated, the majority of invasive procedures are done, and the greatest amounts of antimicrobials are used. The epidemiology, mechanisms of resistance, and strategies used to control and prevent infection by antibiotic-resistant nosocomial pathogens should be understood, with emphasis on multidrug-resistant Staphylococcus aureus, vancomycin-resistant enterococci (VRE), and extended-spectrum, b-lactamase–producing, gram-negative bacilli. 


Following the introduction of each major class of antimicrobials into clinical use, bacterial resistance has emerged. The history of penicillin and resistance in staphylococci is probably one of the best illustrations of emerging resistance. When penicillin was first used in the early 1940s, virtually all staphylococci were susceptible. By 1942, there were a few reports of penicillin resistance in S. aureus isolates; by 1945, resistance rates were as high as 22%.1 By 1967, the rates of resistance in hospitals had increased to more than 80%, and by the 1990s, 90% or more of strains were resistant. Fortunately, semisynthetic penicillins such as methicillin and oxacillin, which were resistant to hydrolysis by staphylococcal b-lactamases, became clinically available in the 1960s, providing an alternative therapy. Resistance to these drugs and other b-lactam antibiotics also developed, however. The prevalence rate for methicillin-resistant S. aureus (MRSA) reported using the National Nosocomial Infection Surveillance System was 5% in 1981 and had increased to 29% in 1991.2 More recent data suggest that this upward trend is continuing, with some large teaching hospitals reporting that 60% or more of S. aureus infections are resistant to methicillin. These methicillin-resistant strains are often resistant to multiple classes of drugs, including macrolides, trimethoprim-sulfamethoxazole, fluoroquinolones, and clindamycin. 

Emergence and Spread

There are a number of factors that contribute to antibiotic resistance. Although it seems intuitive that bacterial resistance develops as a result of exposure to a given antibiotic, the relationship is more complex than that. In general, however, the greater the use of antibiotics, the greater the selective pressures favoring resistant strains. Antimicrobials are the second most commonly prescribed class of drugs in the United States, accounting for as much as 30% to 50% of a hospital’s total drug budget.3,4 Historically, hospitals have been the greatest reservoirs of resistant organisms, partly because of their extensive use of antibiotics; they also congregate patients at risk for infection, including elderly and immunocompromised patients with impaired host defenses. The use of invasive procedures and devices places patients at risk for infection, as well. An estimated 25% to 40% of hospitalized patients receive antimicrobials, with the most use occurring among patients in intensive care units (ICUs).5 The hospital environment also serves as an efficient setting for the spread of resistant organisms from one patient to another, either directly, on the hands of health care workers, or on contaminated objects. 


Bacteria can be resistant to the effects of antibiotics using a variety of mechanisms (Table 1, page 54). One of the most common is production of enzymes that inactivate or modify the antibiotic. Examples of such enzymes include b-lactamases, aminoglycoside-modifying enzymes, and erythromycin esterases. Another mechanism of resistance is decreased accumulation of the drug within the cell. This can result from lessened penetration of the drug (as seen in intrinsic low-level animoglycoside resistance in enterococci) or an active transport system that increases the efflux of the drug, as seen in tetracycline resistance. Alteration of drug target sites—ribosomal proteins, cell-wall precursors, or target enzymes—can also result in resistance. The development of auxotrophs, which are mutants that have growth requirements different from those of the parent strain, allows bacteria to evade the effects of antibiotics. In many instances, multiple mechanisms are involved in the development of bacterial resistance. 


Antibiotic resistance may be an inherent or intrinsic property of a bacterium, or it may be an acquired characteristic. Examples of inherent resistance include the resistance of most gram-negative bacilli to vancomycin, of enterococci to cephalosporins, and of anaerobic bacteria to aminoglycosides. Although clinically important, this type of resistance is of less concern because it is predictable and is usually characteristic of a given species.

Acquired resistance may result from mutations in existing DNA or acquisition of new DNA. Mutations in DNA may be chromosomal, but can also occur in genes residing on plasmids or transposons. Plasmids (extrachromosomal DNA elements that can replicate independently of the chromosome) are probably the most common means of transferring resistance genes in nature. Some plasmids have broad host ranges and can transfer genes between species. An example of this is the transfer by plasmids of genes encoding b-lactamase production and high-level gentamicin resistance from staphylococci to enterococci. Other plasmids have much more restricted host ranges. Transposons, the so-called jumping genes, are DNA elements that can encode for their own movement from one place in the DNA to another. Genes encoding for vancomycin resistance in enterococci may be carried on transposons.

Mutations that lead to antibiotic resistance offer the mutant bacterium a distinct survival advantage in an environment where the antibiotic is present. Antibiotic-resistant bacteria are generally not more virulent or pathogenic than strains of the same bacteria that are not resistant (although, in some instances, genes encoding for virulence factors may be transferred on the same plasmid as resistance determinants). 


In susceptible bacteria, penicillin and other b-lactam antimicrobials act by binding to penicillin-binding proteins (PBPs). These are enzymes in the cell wall that mediate the formation of cell-wall peptidoglycan. This results in disruption of the synthesis of the cell wall and, ultimately, in the death of the bacterium. The major mechanism of methicillin resistance is the production of an altered PBP called PBP2a, which has a low affinity for b-lactam antibiotics.6 An acquired chromosomal gene (mecA) that is not present in methicillin-susceptible strains encodes PBP2a. The origin of mec is unknown, but there is speculation that the genes may have arisen in other species of staphylococci. There is considerable variation in the expression of b-lactam resistance mediated by PBP2a. Heterogeneous resistance, in which only relatively few cells express methicillin resistance, is most common. This presents particular problems because it may be difficult to detect methicillin resistance in these strains in the clinical laboratory. The mechanism by which the expression of methicillin resistance is controlled is not completely understood. A variety of factors or genes have been described that control the expression of methicillin resistance. The relationships and functions of these genes are complex, and further study is needed to understand the regulation of the expression of methicillin resistance. 

The spread of staphylococcal clones carrying the mecA gene (or the horizontal transfer of the gene itself) has resulted in worldwide dissemination of methicillin-resistant staphylococci. Once primarily a problem only in large tertiary care and teaching hospitals, MRSA has moved into smaller community hospitals. Once it has been introduced into a hospital, MRSA is difficult, if not impossible, to eradicate. The principal reservoir of MRSA in hospitals consists of colonized and/or infected patients. Colonization of personnel and persistence of organisms occur, but are of secondary importance. 

Risk factors for MRSA colonization and infection are presented in Table 2. Many of these are also risk factors for the acquisition of methicillin-susceptible S. aureus. As seen for other resistant organisms, prior exposure to antibiotics is an important risk factor. Patients colonized by MRSA are at greater risk for infection by MRSA than patients who are not colonized. 

reduced Susceptibility to Vancomycin 

Recently, S. aureus strains with reduced susceptibility to vancomycin have been reported to cause clinical infections. These strains had vancomycin minimal inhibitory concentrations of 8 mg/mL (intermediate susceptibility) and have been referred to as vancomycin-intermediate S. aureus, or VISA. Glycopeptide-intermediate S. aureus (GISA), however, has generally been the preferred term, since it more appropriately describes resistance to the glycopeptide class of antibiotics (including vancomycin and teicoplanin, a glycopeptide antibiotic available in Europe, but not in the United States). The first reported GISA case occurred in Japan in 1996 and involved an infant with a sternal wound infection.7 Four cases have been reported from the United States.8-10 In each, the patient had received a prolonged course of vancomycin prior to the isolation of GISA strains.

The major significance of these reports is that they portend the emergence of high-level vancomycin resistance in S. aureus. With the widespread appearance of VRE, there has been concern that vancomycin resistance genes will be transferred from enterococci to staphylococci in the clinical environment, particularly since such a transfer has been achieved in the laboratory setting.

The mechanism of resistance in the GISA strains studied to date has not been fully elucidated, but appears to be different from that of vancomycin resistance in enterococci. The genes responsible for vancomycin resistance in enterococci have not been detected in GISA strains. Structurally, the cell walls of GISA strains appear to be different from those of MRSA (which are fully susceptible to vancomycin). Sieradzki et al11 were able to demonstrate that GISA strains grown in broth containing vancomycin were actually able to sequester the vancomycin within the cell wall (presumably, away from the active binding site for the drug), thereby rendering vancomycin ineffective. 


Enterococci have emerged as important nosocomial pathogens over the past decade. Although enterococci are organisms of limited virulence, they are intrinsically resistant to many antibiotics, and they readily acquire resistance genes. They are also capable of transferring resistance genes to other bacteria. The first VRE strain in the United States was isolated in St Louis in 1987. By 1989, strains of VRE had appeared in hospitals throughout the New York City area. In 1997, data collected from United States hospitals indicated that 23.2% of enterococcal isolates from ICU patients and 15.4% from non-ICU patients were vancomycin resistant.12 Although Enterococcus faecalis is more common clinically than Enterococcus faecium, most strains of VRE are E. faecium, and are resistant to ampicillin as well as vancomycin. These strains often demonstrate high-level aminoglycoside resistance, along with resistance to macrolides, fluoroquinolones, tetracycline, and even chloramphenicol. 

VRE can cause a variety of infections. Bacteremias, intra-abdominal infections, and urinary tract infections are, however, most frequent. Although oropharyngeal and upper-airway colonization by VRE may occur, respiratory infections are unusual. Outbreaks of VRE bacteremia, particularly in oncology or transplant units, have been associated with high mortality.13,14 Compared with infections due to susceptible enterococci, VRE infections have been associated with longer stays, greater costs, higher rates of recurrent bacteremia, and greater mortality.15 

A number of risk factors have been identified for colonization and infection by VRE (Table 3, page 54). In almost all studies, prior exposure to antibiotics has been found to be a strong risk factor. Although it is likely that heavy use of vancomycin provided the initial environmental pressure necessary to select for resistant enterococci, other antibiotics—particularly third-generation cephalosporins and drugs with anaerobic activity—appear to be important in maintaining these organisms in the hospital environment. The major reservoir of VRE is the gastrointestinal tract of a colonized patient, where the organism may persist for months. Colonization pressure, or the proportion of patients already colonized by VRE in a given hospital, has been shown to be the major variable affecting acquisition of VRE in hospitals.16 This would suggest that VRE is spread in institutional settings predominantly by cross-acquisition. The greater the number of colonized patients, the greater the chance that health care workers will come into contact with a colonized patient and transmit the organism to another patient. This may be particularly true in the ICU setting. 

VRE strains have been characterized by their resistance phenotypes. Three main phenotypes are of clinical importance.17 VanA strains demonstrate inducible high-level resistance to vancomycin and teicoplanin. Vancomycin resistance in these strains is usually plasmid mediated and can be transferred to susceptible isolates. VanB strains usually have lower levels of resistance to vancomycin and are usually susceptible to teicoplanin. VanC strains are intrinsically resistant to vancomycin (with low or moderate levels of resistance) and remain susceptible to teicoplanin. The VanC phenotype is generally restricted to Enterococcus gallinarium and Enterococcus casseliflavus, species that are rarely pathogenic for humans. 

Vancomycin resistance in VanA and VanB phenotypes is mediated by an acquired cluster of genes that may be carried on plasmids or transposons. Many major outbreaks of VRE have involved VanA strains. Enterococci that acquire these resistance genes produce an altered cell-wall peptidoglycan precursor that does not bind vancomycin, thereby rendering it ineffective. The production of the abnormal cell wall is induced by the presence of vancomycin; that is, the resistance genes are turned on when vancomycin is present. 

Extended-spectrum b-lactamases 

The most important mechanism of resistance to b-lactam antibiotics by gram-negative bacilli is the production of b-lactamase enzymes. Over the past 10 years or more, there has been worldwide emergence of enteric gram-negative bacilli that produce new b-lactamases.18,19 The most important of these are the extended-spectrum b-lactamases (ESBLs). These are a result of amino-acid substitutions in broad-spectrum b-lactamases that extend their substrate profile to include the newer cephalosporins (cefotaxime, ceftriaxone, and ceftazidime) and aztreonam. These organisms are usually multidrug resistant and may also produce such large amounts of b-lactamase that they are able to overcome the effect of b-lactamase inhibitors. They are, therefore, resistant to b-lactamase inhibitor combinations such as piperacillin-tazobactam and ampicillin-sulbactam. In some instances, they may also be resistant to fluoroquinolones, leaving carbapenems (imipenem or meropenem) as the only effective class of antibiotics. Table 4 (page 55) lists the enterobacteriaceae that produce ESBLs. The most frequent ESBL-producing organisms are Klebsiella pneumoniae and Escherichia coli, both important nosocomial pathogens, particularly in ICUs. Hospital outbreaks of ESBL-producing K. pneumoniae have been described worldwide. Prevalence of enterobacteriaceae producing ESBLs in US hospitals has varied from less than 1% to as much as 40%.19 In Europe, where these organisms were first recognized, the prevalence is higher. Risk factors for infection with ESBL-producing organisms are similar to those for other nosocomial pathogens, such as MRSA and VRE (Table 5, page 55). Heavy use of ceftazidime has been implicated in many outbreaks involving ESBL-producing bacteria. The hands of health care workers are likely to be responsible for transmission of the organism from patient to patient. 

The most worrisome aspect of ESBLs may be the difficulties inherent in detecting their presence in the clinical laboratory. Many of these strains are not detected through routine susceptibility testing methods such as agar-disk diffusion and broth microdilution. This may result in erroneous reports of susceptibility to the newer cephalosporins (and subsequent treatment failures). The National Committee for Clinical Laboratory Standards has developed screening criteria for the detection of ESBLs in K. pneumoniae and E. coli.20 These include the testing of the newer cephalosporins, cefpodoxime (an oral cephalosporin) and aztreonam. Cefpodoxime is the most sensitive of these agents for detecting production of ESBL. Strains that produce ESBLs will test intermediately susceptible or resistant to this agent, while strains that do not will be susceptible. Other specialized tests, such as the double-disk diffusion test, are more sensitive and specific in identifying the production of ESBLs. The double-disk diffusion test involves placing a clavulanate disk on a susceptibility plate inoculated with the test organism approximately 30 mm from a disk containing an indicator drug, such as ceftriaxone or ceftazidime. Enhancement of the zone of inhibition between the clavulanate disk and the indicator drug disk indicates the production of ESBL by the organism. There are also commercial systems available for detection of ESBL-producing organisms. 

Prevention and Control strategies 

The prevention and control of bacterial resistance are complex, and they require a multidisciplinary approach. A number of expert task forces and advisory groups have promulgated guidelines and recommendations to reduce antibiotic resistance.21-24 Foremost among these recommendations is controlling and optimizing the use of antimicrobials. This is no easy task, as it requires education and the modification of the behavior and expectations of both physicians and patients. Table 6 (page 55) outlines possible strategies for optimizing the use of antimicrobials, thereby reducing selection pressures. Other recommendations include

• use of improved methods for surveillance of resistant microorganisms, on both national and local levels;
• rigorous adherence to infection control policies and procedures;
• performance of further research to gain a better understanding of the mechanisms of bacterial resistance and spread; and
• development and clinical testing of new antimicrobials. 

Two new antimicrobials, quinupristin-dalfopristin and linezolid, have been approved by the US Food and Drug Administration within the past year for treatment of infection by resistant gram-positive pathogens.25-28 Both of these agents have been demonstrated as safe and clinically efficacious in the treatment of vancomycin-resistant E. faecium infections. Linezolid has also demonstrated clinical efficacy in treating infections due to MRSA. These agents may offer an option where there previously have been no clinically effective drugs, as well as for patients in whom treatment with other agents such as vancomycin has failed (or has not been tolerated).

It is critical that these new antibiotics targeted at multidrug-resistant pathogens (and other agents that follow them) be used judiciously in order to preserve their usefulness. As history has shown only too well, no antibiotic is completely safe from bacterial resistance, and this is something that we would all be wise to remember. N 

Lisa L. Dever, MD, is chief of the Infectious Diseases Clinic, VA New Jersey Health Care System, East Orange, NJ, and associate professor of medicine, UMDNJ-New Jersey Medical School, Newark, NJ.


1. Fishman NO, Brennan PJ. Optimizing use of antimicrobial agents: pitfalls and consequences of inappropriate therapy. Clin Outcomes Management. 1997;4:25-33.

2. Col NF, O’Connor RW. Estimating worldwide current antibiotic usage; report of task force 1. Rev Infect Dis. 1987;9:S232-S243.

3. Maranan MC, Moreira B, Boyle-Vavra S, Daum RS. Antimicrobial resistance in staphylococci. Epidemiology, molecular mechanisms, and clinical relevance. Infect Dis Clin North Am. 1997;11:813-849. 

4. Panlilio AL, Culver DH, Gaynes RP, et al. Methicillin-resistant Staphylococcus aureus in US hospitals, 1975-1991. Infect Control Hosp Epidemiol. 1992;13:582-586. 

5. Gaynes R. The impact of antimicrobial use on the emergence of antimicrobial-resistant bacteria in hospitals. Infect Dis Clin North Am. 1997;11:757-765. 

6. Hartman BJ, Tomasz A. Low-affinity penicillin-binding protein associated with beta-lactam resistance in Staphylococcus aureus. J Bacteriol. 1984;158:513-516. 

7. Hiramatsu K. The emergence of Staphylococcus aureus with reduced susceptibility to vancomycin in Japan. Am J Med. 1998;104:S7-S10. 

8. Centers for Disease Control and Prevention. Staphylococcus aureus with reduced susceptibility to vancomycin—United States, 1997. MMWR Morb Mortal Wkly Rep. 1997;46:765-766. 

9. Centers for Disease Control and Prevention. Update: Staphylococcus aureus with reduced susceptibility to vancomycin—United States, 1997. MMWR Morb Mortal Wkly Rep. 1997;46:813-815. 

10. Centers for Disease Control and Prevention. Staphylococcus aureus with reduced susceptibility to vancomycin—Illinois, 1999. JAMA. 2000;283:597-598. 

11. Sieradzki K, Roberts RB, Haber SW, Tomasz A. The development of vancomycin resistance in a patient with methicillin-resistant Staphylococcus aureus infection. N Engl J Med. 1999;340:517-523. 

12. Martone WJ. Spread of vancomycin-resistant enterococci: why did it happen in the United States? Infect Control Hosp Epidemiol.1998;19:539-545. 

13. Montecalvo MA, Horowitz H, Gedris C, et al. Outbreak of vancomycin-, ampicillin-, and aminoglycoside-resistant Enterococcus faecium bacteremia in an adult oncology unit. Antimicrob Agents Chemother. 1994;38:1363-1367. 

14. Edmond MB, Ober JF, Weinbaum DL, et al. Vancomycin-resistant Enterococcus faecium bacteremia: risk factors for infection. Clin Infect Dis. 1995;20:1126-1133. 

15. Linden PK. Clinical implications of nosocomial gram-positive bacteremia and superimposed antimicrobial resistance. Am J Med. 1998;104:S24-S33. 

16. Bonten MJM, Slaughter S, Ambergen AW. The role of “colonization pressure” in the spread of vancomycin-resistant enterococci: an important infection control variable. Arch Intern Med. 1998;158:1127-1132. 

17. Murray BE. Vancomycin-resistant enterococci. Am J Med. 1997;102:284-293. 

18. Nordmann P. Trends in beta-lactam resistance among Enterobacteriaceae. Clin Infect Dis. 1998;27:S100-S106. 

19. Jacoby GA. Extended-spectrum beta-lactamases and other enzymes providing resistance to oxyamino-beta-lactams. Infect Dis Clin North Am. 1997;11:875-887. 

20. Performance standards for antimicrobial susceptibility testing; eighth informational supplement. Wayne, Pa: National Committee for Clinical Laboratory Standards; 1998. 

21. Shlaes DM, Gerding DN, John JF Jr, et al. Society for Healthcare Epidemiology of America and Infectious Diseases Society of America Joint Committee on the Prevention of Antimicrobial Resistance: guidelines for the prevention of antimicrobial resistance in hospitals. Infect Control Hosp Epidemiol. 1997;18:275-291. 

22. American Society for Microbiology. Report of the ASM task force on antibiotic resistance. Antimicrob Agents Chemother. 1995 (suppl). 

23. Guidelines on the control of methicillin-resistant Staphylococcus aureus in the community. Report of a combined working party of the British Society for Antimicrobial Chemotherapy and the Hospital Infection Society. J Hosp Infect. 1995;31:1-12. 

24. Centers for Disease Control and Prevention. Recommendations for preventing the spread of vancomycin resistance. MMWR Morb Mortal Wkly Rep. 1994;44:1-13. 

25. Linezolid (Zyvox). Med Lett Drugs Ther. 2000;42:45-46. 

26. Clemett D, Markham A. Linezolid. Drugs. 2000;59:815-827. 

27. Winston DJ, Emmanouilides C, Kroeber A, et al. Quinupristin/dalfopristin therapy for infections due to vancomycin-resistant Enterococcus faecium. Clin Infect Dis. 2000;30:790-797. 

28. Lamb HM, Figgitt DP, Faulds D. Quinupristin/dalfopristin: a review of its use in the management of serious gram-positive infections. Drugs. 1999;58:1061-1097.