Antimicrobial Chemotherapy, 5th Edition

Part 2 - Resistance to Antimicrobial Agents

Chapter 11

Control of the Spread of Resistance

The 60-year period during which antibiotics have been available has seen dramatic changes in the disease burden caused by infections. Outcomes from infections such as pneumococcal pneumonia, tuberculosis, and streptococcal puerperal sepsis, that used to cause considerable morbidity and mortality, are now frequently benign, at least in developed countries. We can also prevent much infection by using antibiotics during high-risk procedures, notably in the peri-operative period. The immense social, economic, and health benefits that are due to antibiotic use are, however, increasingly overshadowed by the issue of resistance. Indeed, the emergence and spread of multiresistant strains (sometimes referred to emotively as ‘superbugs’) have raised the spectre of untreatable infection. The reality is that such instances remain extremely rare. However, resistance does limit antibiotic choice available to prescribers, sometimes meaning that less effective, more toxic or more expensive drugs have to be used. For example, the antibiotics needed to treat multiresistant forms of tuberculosis are over 100 times more expensive than the first-line drugs used to treat disease caused by fully susceptible strains. Such excess costs mean that some infections can no longer be treated in poor communities where resistance to first-line drugs is widespread. Furthermore, significant slowing in the development of genuinely new antibiotics (i.e. those with novel modes of action to which cross-resistance to older agents does not occur) has increased the potential for this threat to become a reality that once again compromises patient outcome.

In 1945 during his Nobel Prize acceptance speech Sir Alexander Fleming said ‘It is not difficult to make microbes resistant to penicillin in the laboratory by exposing them to concentrations not sufficient to kill them, and the same thing has occasionally happened in the body.’ This warning was evident less than a decade after the introduction of penicillin, when a particular penicillin-resistant Staphylococcus aureusstrain started to cause outbreaks of postoperative and perinatal infection in hospitals across the world. Poor hospital cleaning, increasing dependence on antibiotics and changing healthcare practices were blamed. Unfortunately, these issues are again topical, with frequent media headlines about ‘superbugs’ and their spread.

Compared with most other drugs of similar potency, antibiotics are remarkably safe, and they are also remarkably effective. This has inevitably led to liberal, even lavish use, and concern has frequently been expressed that excessive and inappropriate use of these agents is the chief cause of the widespread emergence of resistant organisms. Misuse of most drugs tends to have consequences only for the individual patient. Unfortunately, inappropriate antibiotic use can have adverse consequences for both the individual and for wider populations. Figure 11.1 shows the disturbing relationship between the prescribing of penicillin-like antibiotics in multiple populations and the respective prevalence of pneumococcal strains with reduced susceptibility (or frank resistance) to penicillin. Of course, many of the antibiotics prescribed would not have been specifically for pneumococcal infection. This is, therefore, evidence of the selective pressure for resistance emergence in (respiratory tract) flora and subsequent spread of bacteria within populations. Such effects have been referred to as the collateral damage associated with antibiotic therapy.

Availability of antibiotics

Most developed countries have tightly regulated systems for the control of the manufacture, importation, distribution, sale, supply and description of medicinal products, including antibiotics, for human and veterinary use (see Chapter 32). In the global market for medicines, licensing authorities will increasingly be required to ensure that there is a consistency of approach to medicines availability. Currently there are many examples of inconsistencies in the availability and recommendations for use of antibiotics throughout both the developed and developing world. The availability of antibiotics, notably newer more expensive agents is an issue in poorer countries. Pharmaceutical companies have a part to play in helping to ensure that antimicrobial agents, including critical antimalarial, antituberculosis, and antiretroviral drugs are priced and advertised appropriately in these markets.

While the sale and distribution of antibiotics are fairly tightly controlled in rich, developed countries, the marketing of these agents is much less restricted in the poorer, and numerically much larger, developing world. Paradoxically, the use of antibiotics in the developing countries needs to be extended, not restricted, if standards of health are to be brought up to those of the developed world. A key issue here is unregulated ‘over the counter’ availability of antibiotics. Controversy persists about striking a balance between making effective medicines available in a timely fashion to those who need them, against the potential detrimental effects of uncontrolled or indiscriminate use. This argument is most pertinent in the case of antimicrobial drugs, as there is no other example in therapeutics in which local misuse of an efficacious agent can lead to a general diminution in its effectiveness. It was no great surprise that chloramphenicol-resistant typhoid bacilli first emerged in South America and penicillin-resistant gonococci in south-east Asia, where unrestricted availability of antibiotics is commonplace. In some countries antibiotics can still be purchased easily as single tablets resulting in inappropriate use, suboptimal dosing and the consequent encouragement of resistance.


Fig. 11.1 Correlation between antimicrobial use (outpatient prescribing of penicillins) and resistance (prevalence of penicillin-non-susceptible Str. pneumoniae) in 19 countries in Europe. Reprinted from: Goossens H, Ferech M, Vander Stichele R, Elseviers M. Outpatient antibiotic use in Europe and association with resistance: a cross-national database study. Lancet 2005; 365: 579-587 with permission from Elsevier.


In the UK fluconazole and aciclovir have been available for over-the-counter purchase without the need for a prescription for more than a decade. There is no convincing evidence that this availability additional to prescribed courses has increased the emergence of resistance to these agents in the target pathogens—Candida albicans and herpes simplex virus—for which they are commonly used. This may, however, reflect inherent properties of these drugs uncommonly to select for resistant variants. There is pressure to extend the availability of over-the-counter antibiotics to include drugs such as trimethoprim for use in urinary tract infections. It will be important to monitor any such changes closely to determine the benefits and drawbacks of any such deregulation of antibiotics. A related issue is the extension of capacity to prescribe antibiotics (and other drugs) to other healthcare professionals, including pharmacists and nurses. Such extended roles clearly need to be underpinned by appropriate training and education, and the availability of carefully constructed guidelines (see below).

Inappropriate antibiotic use

Attention has repeatedly been drawn to the worldwide public health problem of the spread and persistence of drug-resistant organisms, and there have been frequent calls for regulation to curb the unnecessary use and misuse of antimicrobial drugs in some countries. The following practices have been clearly identified as contributing to the present situation:

  • Inappropriate prescribing of antibiotics;e.g. for ailments for which they are ineffective, such as for sore throats where 80% of infective cases are caused by viruses (see Chapter 19).
  • Incorrect dose or duration of use;e.g. in uncomplicated urinary tract infection more than 3 days of antibiotic treatment does not increase the chance of success, but does increase the risk of selection of resistance bacteria in the gut flora and adverse drug effects (see Chapter 20).
  • Excessive use of antibiotic prophylaxis;e.g. for most types of surgery there is no value in giving more than one dose of antibiotic(s)

(See Chapter 17). Excess antibiotic doses may encourage resistance emergence or side effects including antibiotic-associated diarrhoea.

  • Antibiotic use without prescription;e.g. the uncontrolled availability of antibiotics ‘over the counter’, which can result in unnecessary use or intermittent, suboptimal dosing. The increasing availability of antibiotics through the internet may exacerbate this risk. In some countries, poorly formulated or manufactured, counterfeited or expired antibiotics are sold and used for self-medication or prophylaxis.
  • Animal/agricultural use of antibiotics;e.g. using clinically useful antibiotics as growth promoters in animal feeds and on agricultural crops (see below).

Antibiotics use in animals

More than half of all antibiotics produced worldwide are used in animals, primarily as part of the food production chain. Two aspects of this use are particularly concerning. First, there is a large overlap between the types of antibiotics given to animals and those used to treat infection in man. Secondly, large quantities of antibiotics are used not to treat overt infection but instead as animal growth promoters to increase weight gain and therefore market value of animals. Combining these two issues, it is not surprising therefore that there is mounting evidence of resistant bacteria developing in animals and either infecting human beings or acting as a source of resistance genes for human pathogens. For example, avoparcin use in animals is linked to the development of resistance to glycopeptides in animal strains of enterococci and possibly also in human strains. Avoparcin was banned as a growth promoter in Denmark in 1995, at which point about 80% of Danish broiler chickens were colonized with vancomycin-resistant enterococci; the current prevalence is less than 5%. Similarly, fluoroquinolone use in animals has been clearly associated with the increase in prevalence of fluoroquinolone resistance in salmonella and campylobacter strains that infect man. Notably, a multiresistant Salmonella enterica serotype Typhimurium strain (DT104) has spread in animals, foods and, subsequently, in human beings.

All use of antimicrobial agents for growth promotion is now banned in the European Union. There is some concern that the therapeutic use of antibiotics in animals may increase as use of antibiotic growth promoters is curtailed, but this is unlikely to have the same negative consequences as seen with unrestricted use of antibiotics in animals.


Antibiotic prescribing in the community and hospital

The European Union, the US Food and Drug Administration and the World Health Organization have initiated national and regional campaigns aimed at professionals and the public to reduce the unnecessary prescribing of antibiotics. Efforts have been concentrated on prescribing in the community, not least because this accounts for 80% of all human use of antimicrobial drugs. Principles such as not prescribing antibiotics for viral sore throats, or simple coughs and colds, and avoiding the use of new and more expensive antibiotics (e.g. quinolones and cephalosporins) when standard and less expensive antibiotics remain effective have been emphasized. Prescribing of antibiotics started to fall in England in 1995-1996. The decrease subsequently stabilized, with a slight rise in 2003-2004 (Fig. 11.2).

It is not clear whether this decrease in prescribing has been driven by reduced incidence of infections (such as respiratory tract infections), reduced consultation rates, or because general practitioners are following prescribing guidance for infections more closely. However, prescribing of paediatric antibiotic preparations fell by almost 50%—a much greater reduction than that seen for the whole population. This suggests that public (parental) expectation, and hence pressure, for the doctor to prescribe an antibiotic following a consultation may be decreasing.


Fig. 11.2 Trends in prescribing of antibacterial drugs in general practice in England. Reproduced from: National Health Service Business Services Authority prescription pricing division website. Available at:


It is estimated that up to 50% of antibiotic usage in hospitals is inappropriate. Interventions to improve antibiotic prescribing for hospital inpatients can be successful, and importantly may reduce antimicrobial resistance or hospital-acquired infections, such as Clostridium difficile infection. A key issue is choosing the most appropriate control methods for a given setting and ensuring that they are sustainable. The scope of measures that can be used to reduce inappropriate antibiotic prescribing is too large to consider in detail here, but can generally be grouped into educational or restrictive approaches or a combination of the two.

Appropriate antibiotic use

The World Health Organization advocates the following 12 key interventions to promote more rational use of medicines in general; all are applicable to antibiotic use:

  • Establishment of a multidisciplinary national body to coordinate policies on medicine use.
  • Use of clinical guidelines.
  • Development and use of national essential medicines list.
  • Establishment of drug and therapeutics committees in districts and hospitals.
  • Inclusion of problem-based pharmacotherapy training in undergraduate curricula.
  • Continuing in-service medical education as a licensure requirement.
  • Supervision, audit, and feedback.
  • Use of independent information on medicines.
  • Public education about medicines.
  • Avoidance of perverse financial incentives.
  • Use of appropriate and enforced regulation.
  • Sufficient government expenditure to ensure availability of medicines and staff.

In addition to these measures, good antimicrobial prescribing needs to be informed by timely and accurate information on the likely infecting pathogens. Delays in diagnosis occur through poor or non-existing sampling techniques, delay in transport, slow and laborious laboratory techniques, and unsatisfactory reporting methods (see Chapter 12). A major problem in dealing with patients in whom an infection is suspected is distinguishing between infection and colonization. Patients with an undiagnosed fever may well be colonized with potentially pathogenic micro-organisms, but may not be infected. The distinction is not always obvious, and under these circumstances it is understandable for a clinician to prescribe antibiotics. It is not rational, however, to treat patients merely because they have a raised temperature. Good practice dictates that all relevant samples for culture should ideally be collected before treatment, unless this requirement could compromise outcome (for example, in patients with suspected meningitis where prompt antibiotic therapy may be life saving). The initial choice of antimicrobial therapy will depend on the most likely infecting organism, the severity of the illness, and the type of patient (see Chapter 13). If the identity of the organism is known then treatment can be specific and a single, narrow-spectrum antibiotic used. If the infecting organism can be targeted then broad-spectrum antibiotics do not need to be used, thus leaving much of the body's normal flora undisturbed.

Initial (often empirical) antibiotic therapy is based on good surveillance and prompt guidance informed by accessible policies (see Chapter 18) or from infection specialists. Crucially, antibiotic prescriptions should be reviewed regularly to determine whether the drug or route of administration is still appropriate. Oral antibiotics tend to be considerably cheaper than intravenous alternatives and of course do not require an access device that itself may be a source of infection. For these reasons, intravenous antibiotics should be reviewed after 48-72 h and switched to an ‘equivalent’ oral formulation, provided oral absorption is satisfactory and the oral antibiotic has the requisite pharmacokinetic characteristics. New microbiological or other information (e.g. fever defervescence for at least 24 h, marked clinical improvement; low C-reactive protein) should prompt a review of therapy and consideration of whether a switch to oral antibiotic(s), a narrow spectrum intravenous alternative, or cessation of antibiotics (no infection present) is appropriate. Laboratory reports should contain information on a restricted number of antibiotic susceptibilities (see Chapter 12).

Antibiotic policies and resistance surveillance

The principles that should be followed in deciding which antibiotic, if any, to use in a given situation are discussed in Chapter 13. However, even in relatively straightforward clinical situations there are often several equally effective agents that might be used. Choice may then be determined by a locally agreed set of guidelines for the rational use of antibiotics. The antibiotic formulary is a locally agreed list of available antibiotics, usually including some degree of restriction on particular agents. Guidance on the most appropriate use of antibiotics should not be too restrictive, should reflect local needs, and should be formulated with the agreement of the local users. Advice should of course facilitate the most effective treatment for the individual patient, but should take into account the potential consequences for the wider population. These issues are considered more fully in Chapter 18.

The best antibiotic policies are grounded in good microbiology laboratory surveillance, which is required to detect important change in bacterial resistance. Clinicians need to be aware of the local and changing patterns of infection and antibiotic resistance in their locality. Information about new agents, together with an assessment of their likely place in therapy, should be available. There are a number of caveats to pathogen and antibiotic surveillance data in general. Bias inherent in the way samples or pathogens are collected is a common problem. For example, uncomplicated urinary tract and respiratory tract infections are usually treated empirically and indeed without samples being submitted. General practitioners tend to reserve the submission of urine or sputum samples for those cases that have complicated courses or where recurrence of symptoms occurs. Thus, antibiotic treatment policies based entirely on the results of such samples and pathogens will tend to be skewed towards more antibiotic-resistant pathogens, and in turn may recommend unnecessarily broad spectrum or newer antibiotics. Such issues can be overcome by using sentinel (sometimes also called spotter) practices that submit samples from patients with ‘normal’ infections, usually for set periods of the year.

Antibiotic rotation

The selective pressure that results from relying on one or a few antibiotics has led some to explore whether antibiotic rotation (also called antibiotic cycling), particularly in the intensive care unit, can reduce or delay the emergence of resistance. However, this theory has several important unanswered issues: how often should antibiotics be rotated? Is the optimum period of usage the same for all antimicrobial drugs? Which antibiotics and classes should be rotated and in what order? What are the practicalities of ensuring compliance with a rotational policy? Current consensus is that routine antibiotic rotation should not be implemented. Indeed, several studies have been unable to demonstrate a reduction in the prevalence of resistance while antibiotic rotation was being used, and some have found that resistance actually increased during some parts of the cycle. Ironically, diverse antimicrobial prescribing may be associated with reduced emergence of resistance. This should not be interpreted as an argument for entirely unrestricted prescribing, as this is likely to be associated with suboptimal therapy for some patients.

Monitoring antibiotic policies

Blind faith in a restrictive antibiotic policy is not the answer to control of antibiotic usage since bacterial resistance patterns change over time owing to selective pressure. Periodic antibiotic audit should be mandatory in all areas where prescribing occurs. This should not be viewed as a policing exercise, so implying a threat to the clinician's freedom to prescribe, but instead should serve as a need to justify selection of antimicrobial agents in the light of critical analysis. Monitoring antibiotic usage should provide ward, unit, and hospital-wide information on prescribing patterns. This should prove useful for trend analysis and allow discrepancies to be identified. Such information lends itself to detailed scrutiny to differentiate between rational, questionable, and irrational antibiotic usage. Clinical efficacy and adverse events can be evaluated. Correlations between antibiotic usage and antimicrobial resistance should be sought, and changes can be made. A ‘defined daily dose’ for each antibiotic can be used as a standard unit of measurement, and can be useful to identify qualitative as well as quantitative variability in prescribing.

The next stage on from monitoring antibiotic usage is antibiotic audit, thereby closing the audit loop (Table 11.1).

Control of the transmission of antibiotic-resistant bacteria

It is essential that an active infection control programme is also in place, so that patients harbouring multiresistant bacteria are appropriately nursed, managed, and treated. While a full account of the optimal infection control procedure to minimize the risk of pathogen transmission is not appropriate here, some important principles are worth emphasizing. Isolation of patients and ensuring scrupulous hand hygiene, such as with alcohol-based hand rubs, can reduce the risk of transmission and the spread of pathogens. Much has been written about hospital cleanliness and the risk of hospital infection, notably the spread of antibiotic-resistant bacteria such as methicillin-resistantStaph. aureus (MRSA), but the lack of data to substantiate a link between these is stark. The great majority of infections acquired during healthcare arise because of poor hand hygiene. Compliance with hand hygiene policies should therefore be monitored. Although the need to isolate a patient may conflict with other pressures on healthcare delivery, this should not prevent infection control teams implementing this fundamental way of minimizing pathogen dissemination risk where appropriate. Patients may be isolated in single rooms or cohort-isolated in groups of beds or on dedicated units.

Table 11.1 Audit of antibiotic prescribing


What was the most likely anatomical site of the infection?


What clinical and radiological evidence, and laboratory findings supported the evidence of infection?


Were bacteriological investigations requested?


Was a pathogen identified?


If not, what was the suspected pathogen?


Were there any recognized underlying or predisposing conditions relevant to the diagnosis of infection?


Was the diagnosis, confirmed, doubtful, or unknown?


Should the patient have received an antibiotic?


Was the antibiotic an appropriate choice?


Was the dose appropriate?


Was its frequency of dosage appropriate?


Was the route of administration appropriate?


Was the course of treatment of suitable duration?

Much infection control practice is based on empiricism, and policies are frequently based on experience rather than controlled trial data. This does not mean that such policies are optional! There is ample evidence that when infection control measures are strictly enforced, the incidence of infection with resistant organisms can be reduced. Effective ways of preventing cross-infection with and spread of antibiotic-resistant pathogens still need to be defined and refined. Crucially, these approaches may need to differ depending on whether a particular antibiotic-resistant pathogen has already become established (endemic) or is rare (sporadic). A good analogy is plugging the holes in a leaking dyke: eventually more than fingers are needed to sustain the barrier. Controversy still exists about the true control benefit of screening for specific potential pathogens such as MRSA. The role and benefit of new rapid screening methods, usually based on DNA detection, remain to be determined. Alternative approaches include targeted prophylaxis against such pathogens in patients undergoing high-risk procedures such as surgery.


Further reading

Davey P, Brown E, Fenelon L, Finch R, Gould I, Hartman G, Holmes A, Ramsay C, Taylor E, Wilcox M, Wiffen P. Interventions to improve antibiotic prescribing practices for hospital inpatients. The Cochrane Database of Systematic Reviews 2005, Issue 4. Art. No.: CD003543. DOI: 10.1002/14651858.CD003543.pub2. Available at:

World Health Organization. WHO global strategy for containment of antimicrobial resistance.