Antibiotic resistance is a growing problem in New Zealand
Resistance to commonly used antimicrobial medicines, in a wide range of bacteria responsible for common diseases, is rapidly emerging as a major threat to health in New Zealand. For example, in New Zealand during 2012, approximately 4,000 people suffered from infections due to strains of Escherichia coli (E. coli) or Klebsiella pneumoniae (K. pneumoniae) or other closely related organisms that produced an extended spectrum beta-lactamase (ESBL).1
The production of an ESBL by these bacteria confers resistance to almost all penicillins and all cephalosporins. Associated resistance genes in these ESBL-producing bacteria frequently confer resistance to many other antimicrobials.
The epidemic of ESBL-producing bacteria in New Zealand, has grown rapidly in recent years. During 2006–2008 approximately 2.6% of E. coli isolated from blood cultures in New Zealand were ESBL positive; by 2009–2011 the proportion had risen to 3.8%—a 46% increase.2
Experience in other countries suggests that ESBL-producing bacteria may become much more common in New Zealand in the next few years. For example, between 2000 and 2011, the proportion of isolates of E. coli that were resistant to third-generation cephalosporins (an approximate surrogate for the production of an ESBL) increased from 0.1% to 5.7% in the Netherlands, and from 3.6% to 14.9% in Greece.3
The dramatic increase in ESBL-producing E. coli and K. pneumoniae is but one example of the growth in antimicrobial resistance in New Zealand in recent years. Resistance to mupirocin, the active component in Bactroban© ointment, was present in <5% Staphylococcus aureus strains isolated from patients in Auckland in 1992, but by 2000 had risen to >20%.4 The proportion of strains of Neisseria gonorrhoeae (N. gonorrhoeae) that are resistant to ciprofloxacin has risen from 6% in 2002 to 40.6% in 2012.5 (Figure 1)
Figure 1. The prevalence of resistance to ciprofloxacin in isolates of N. gonorrhoeae tested by community and hospital laboratories in New Zealand between 2002 and 20125
Increasing levels of antibiotic resistance cause inconvenience and risk for patients and increased costs for the health system
The emergence of high rates of resistance, in a wide range of bacterial species, during recent decades, has had significant impacts on patients and on the healthcare system. For most of the approximately 4000 patients who suffered from an infection due to an ESBL-producing E. coli or K. pneumoniae during 2012, the only reliable and safe treatment would have been a carbapenem antibiotic, such as meropenem or ertapenem. Because these antibiotics are not orally absorbed, patients with infections due to an ESBL-producing strain of E. coli or K. pneumoniae commonly require intravenous therapy, usually administered within hospital.
In contrast, patients infected with strains of E. coli or K. pneumoniae that do not produce an ESBL can commonly be treated with an oral agent, such as trimethoprim, amoxicillin plus clavulanate, ciprofloxacin or nitrofurantoin, and therefore usually do not require admission to hospital. Because of the high prevalence of resistance to ciprofloxacin, patients with gonorrhoea are now usually treated with intramuscular ceftriaxone, instead of oral ciprofloxacin.
Use of second-line antibiotics to treat these more resistant strains of E. coli, K. pneumoniae, and N. gonorrhoeae results in increased discomfort and inconvenience for the patient, and increased costs for the healthcare system.
Completely untreatable bacterial infections are a rapidly emerging threat
At present, for most infections due to antibiotic-resistant bacteria, a relatively effective treatment is available, even if this treatment has disadvantages with regard to cost or convenience or adverse effects. However strains of E. coli and K. pneumoniae that produce a carbapenemase, and so cannot be effectively treated with meropenem or ertapenem, have been isolated from relatively small numbers of patients in New Zealand,6 and there has been an alarming increase in their prevalence in hospitals overseas. These infections are often untreatable and cause high death rates.7
A similar problem has recently emerged in Russia where excessive antimicrobial consumption has led to resistance to all available antimicrobial agents in over 5% of strains of Pseudomonas aeruginosa (P. aeruginosa) isolated from hospital inpatients.8
If untreatable infections, due to E. coli, K. pneumoniae, P. aeruginosa or other completely drug-resistant organisms become common here, this will have major consequences for the New Zealand healthcare system. For example, the surgical implantation of prosthetic materials, and the potent immunosuppression required for successful solid organ or bone marrow transplantation, place many patients at risk of severe infection by a wide range of organisms, including E. coli, K. pneumoniae and P. aeruginosa.
At present, because of the very low rates of completely drug-resistant organisms in New Zealand, the infectious risks of these modern medical procedures are considered acceptable for most patients. However, if an increasing proportion of procedure-related infections are caused by untreatable bacteria, the infectious risks may be considered much less acceptable and the use of these procedures curtailed. Increasing antibiotic resistance therefore threatens a very wide range of current medical and surgical practices. It should be of great concern to the community and to all healthcare workers.
New antibacterial drugs will not be available in the next decade
Until about 15 years ago the emergence of resistance to a single antimicrobial, or even to all members of a class of antimicrobials, was merely an inconvenience.
During the 1970s and 1980s, large numbers of new antimicrobials had been developed, and together these new drugs ensured that infections due to almost any bacteria could be readily cured. However, since the 1990s, pharmaceutical companies have devoted minimal resources to development of new antimicrobials and the pipeline of new drugs has dried up.9
It seems very unlikely that new drugs will be developed in time to effectively treat infections caused by the increasing numbers of highly resistant bacteria. Consequently we must change our strategy from accepting the emergence of bacterial resistance as a minor inconvenience to doing all that is feasible to slow the emergence of resistance.
High levels of consumption of antibiotics is the primary cause of high rates of antibiotic-resistant bacteria
The emergence and proliferation of bacteria resistant to an antibacterial is directly related to the amount of that antibacterial in the organism’s environment.
As a result, the prevalence of strains of bacteria resistant to an antimicrobial class increases more rapidly in those countries where large amounts of antimicrobials within that class are consumed. For example, strains of S. pneumoniae with reduced susceptibility to penicillins are common in Spain and France, where large amounts of penicillins are consumed.3,10 In contrast, strains of S. pneumoniae with reduced susceptibility to penicillins are unusual in Germany and the Netherlands, where lower amounts of penicillins are consumed (Figure 2a).
Similarly strains of E. coli resistant to ciprofloxacin are common in Spain and Italy, where large amounts of fluoroquinolones are consumed.3,10 In contrast, strains of E. coli resistant to fluoroquinolones are less common in Norway, Sweden and Finland, where lower amounts of fluoroquinolones are consumed (Figure 2b).
Figure 2. Consumption of penicillins (defined daily doses [DDDs]/1000 population/day) by community-based patients (i.e. not hospital inpatients) during 2010, in relation to the prevalence of reduced susceptibility to penicillin in strains of S. pneumoniae isolated during 2010; (a) and consumption of fluoroquinolones (DDDs/1000 population/day) by community-based patients during 2010, in relation to the prevalence of resistance to ciprofloxacin in strains of E. coli isolated during 2010; (b) for a number of large European nations.3,10
Note: The per capita level of antimicrobial consumption by community-based patients is commonly measured in defined daily doses (DDDs) per 1000 population per day. The DDD is the weight of each medicine that has been agreed by an international panel to be the standard daily dose used when treating an otherwise healthy adult (e.g. oral amoxicillin: 1g; roxithromycin: 300mg; ciprofloxacin: 1g; doxycycline: 100mg; trimethoprim: 400mg).11
Antibiotic consumption in New Zealand has risen rapidly and is high by international standards
Antimicrobial consumption in New Zealand has been relatively prudent in the past, and consequently the prevalence of antimicrobial resistance in most organisms has, until recently, been relatively low.
However, in New Zealand during the 7 years between 2005 and 2012, annual per capita antimicrobial consumption by community-based patients increased by 43%, an average annual increase of just over 6% (Figure 3).
Figure 3. Annual per capita consumption of antimicrobials, by community-based patients in New Zealand, measured in DDDs/1000 population/day, between 2005 and 2012
Note: Community pharmacy dispensing data for antimicrobials was obtained from the National Pharmaceutical Collection, in the Ministry of Health;12 and the number of people resident in New Zealand during each year was obtained from Statistics New Zealand.13
The overall level of consumption of antibiotics, by community-based patients in New Zealand in 2010, was less than that in Greece, Belgium, France and Italy, but was greater than that in Spain and most other European countries (Figure 4).10
The level of antimicrobial consumption in New Zealand in recent years is most comparable with those European countries that are widely considered to have profligate levels of antimicrobial consumption, and that in consequence have high levels of antimicrobial resistance.
Figure 4. Annual per capita consumption of antimicrobials by community-based patients, in various European countries 10 and in New Zealand, during 2010, measured in DDDs/1000 population/day
Programs to reduce antibiotic consumption in New Zealand are urgently needed
New Zealand needs to promptly institute a range of effective measures to reduce antimicrobial consumption. In the absence of such measures the prevalence of resistance in common pathogens will rise quickly and we will suffer the consequences outlined above.
Since 1999 New Zealand’s Pharmaceutical Management Agency (PHARMAC) has funded and managed the Wise use of antibiotics campaign that was associated with a dramatic decline in prescriptions for amoxycillin plus clavulanate during 1999 and 2000,14 and that probably also contributed to the slower rate of growth in overall consumption since 2009 (Figure 3).
The slower rate of growth of antimicrobial consumption in recent years may also be a result of increased prescriber compliance with the guidance provided by the Best Practice Advocacy Centre (BPAC)15 and attention by prescribers to the feedback from BPAC about their level of antimicrobial prescribing. However, despite these efforts to encourage prudent antimicrobial prescribing in New Zealand, the per capita consumption of antimicrobials has grown, rather than stayed level or declined.
Sustained reductions in antimicrobial consumption in New Zealand in the coming years will require recognition that this is a goal with major long-term benefits. Reducing the level of antimicrobial consumption is analogous to increasing the level of infant immunisation: effort expended now is more than repaid in the future.
In 2009, the Ministry of Health recognised the potential benefits of setting targets for the uptake of childhood immunisations, and regularly reporting the performance of DHBs in relation to these immunisation targets.16
Subsequently there has been a significant improvement in immunisation uptake in all population groups throughout New Zealand. In a similar manner the Ministry of Health now should set targets for reductions in the per capita antibiotic consumption and regularly report the performance of DHBs in relation to these targets.
Figure 5 shows that the annual per capita consumption of antibiotics by community-based patients during 2012, (measured in DDDs/1000 population/day) varied between 18.93 in the West Coast DHB, and 28.13 in Counties Manukau DHB.
A number of factors, such as the proportions of the population comprised by children, and by people experiencing socioeconomic deprivation, will cause some variation between DHBs in the prevalence of infections that require antibiotic treatment.
Therefore it is not realistic to expect that the per capita level of antimicrobial consumption will be equivalent for all DHBs. However, it is inevitable that those DHBs with higher levels of antimicrobial consumption will experience more rapid growth in antimicrobial resistance than those DHBs with lower levels of consumption.
As a result, in the near future, DHBs with consistently high levels of antimicrobial consumption will more commonly confront the problem of patients with infections due to resistant organisms. Therefore, it should be a goal for all DHBs to reduce unnecessary antimicrobial consumption and slow the emergence of antimicrobial resistance within their population.
Figure 5. Annual per capita consumption of antimicrobials by community-based patients, in the 20 New Zealand District Health Boards, during 2012, measured in DDDs/1000 population/day
Strategies to reduce antimicrobial consumption in New Zealand
In developed countries, a large proportion of antimicrobial treatment, often more than 50%, is prescribed for patients with conditions in which antimicrobial treatment makes no significant impact on mortality or morbidity. For example, despite widespread understanding that antimicrobials provide no benefit for patients with viral upper respiratory tract infections, it is disappointingly common practice in New Zealand for patients presenting with symptoms suggestive of such illnesses to be prescribed an antimicrobial.17,18 Other conditions for which antimicrobial treatment is commonly prescribed, with nil or negligible benefit, include otitis media, boils and most diarrhoeal illnesses.15
Ideally treatment of infections should be as narrow spectrum as possible. For example, penicillin should be preferred over amoxicillin, cephalexin or amoxicillin/clavulanate in the treatment of Streptococcus pyogenes or S. pneumoniae infection, and flucloxacillin should be preferred over cephalexin or amoxicillin/clavulanate in the treatment of Staphylococcus aureus infection.
Antimicrobial treatment should be prescribed for the shortest duration necessary to achieve a significant impact on mortality or morbidity, and not prolonged unnecessarily.
Prescribers may be reluctant to select the most narrow spectrum antimicrobial agent for the treatment of infections, because other, more broad spectrum, agents may have advantages with respect to the palatability of the medicine or the convenience of the dosing regimen. For example, amoxicillin may be preferred over penicillin because there is no requirement for the medicine to be taken well away from meals, or cephalexin may be preferred over flucloxacillin because it can be taken twice, rather than four times, daily.
The overall benefits and disadvantages of the various antimicrobials that might be used for the treatment of an infection should be carefully considered. The contribution of broad spectrum agents to speeding the emergence of antimicrobial resistance should not be ignored, or considered trivial in comparison to other considerations such as palatability or convenience.
Narrow-spectrum penicillins such as penicillin V and flucloxacillin are under-prescribed in New Zealand when compared with other more prudent countries. During 2012, consumption of narrow spectrum penicillins comprised only 21% (2.73/13.22 DDDs/1000 population/day) of total consumption of penicillins by community-based patients in New Zealand.
In contrast, during 2012, consumption of narrow spectrum penicillins comprised 57% and 77% of total consumption of penicillins by community-based patients in Denmark and Sweden. 10 In these countries a much greater emphasis is placed on the use of narrow spectrum antimicrobials to slow the emergence of resistance.
Figure 6 shows that in New Zealand, as in other countries, the level of antimicrobial consumption is highest in children aged less than 5 years and in adults over the age of 80 years. However there was no age group that was dispensed less than 60 antibiotic prescriptions per 100 population. This is consistent with a recent study of antibiotic consumption in Tairawhiti which found that, during a 12 month period in 2005 to 2006, at least one antibiotic prescription was dispensed to 60% of males and females aged 0–15 years, 33% of males and 48% of females aged 35–44 years, and 65% of males and 70% of females aged ≥85 years.19
These data suggest that strategies to reduce antimicrobial consumption in New Zealand should not be focused exclusively on paediatric prescribing but rather should be designed to influence prescribing of antibiotics across all age groups.
Figure 6. Per capita consumption of antimicrobials by community-based patients, in relation to patient age, during 2012, measured in prescriptions/100 population/year.
Note: Data for the number of antimicrobials dispensed by community pharmacies during 2012, for patients within each age range, was obtained from the National Pharmaceutical Collection;12 and for the number of people within each age range resident in New Zealand during 2012 was obtained from Statistics New Zealand.13
Programs that encourage reduced antimicrobial consumption are effective
Various measures to reduce antimicrobial consumption, and thus slow the increase in the prevalence of resistance, have been used overseas.
In 2002, France conducted a nationwide campaign Antibiotics are not automatic that over the subsequent 5 years resulted in a greater than 25% per capita reduction in antibiotic prescriptions. 20
In 2010, the Swedish Government set a target to reduce total antimicrobial consumption by 36% over 4 years, from 39 antibiotic prescriptions per 100 population per year, to less than 25 antibiotic prescriptions per 100 population per year.21
Systematic reviews suggest that a variety of educational strategies targeted at the prescribing doctor and/or their patients can result in an average 25% reduction in the proportion of patients who are prescribed an antibiotic.22 There is little evidence that one strategy is more effective than others, however “passive” education of prescribers appears to have relatively little effect and encouraging use of “delayed dispensing” appears to have a relatively large effect.23
Patients should be educated about the adverse effects of antibiotic consumption
Patients commonly appreciate the dramatic benefits of antibiotic treatment of serious bacterial infections, but commonly are unaware of the potential harms of antimicrobial consumption. As a result patients may unrealistically exaggerate the benefits and neglect the disadvantages of antibiotic treatment for a wide range of relatively minor illnesses.
One disadvantage of an unnecessary antimicrobial treatment that patients may neglect is the increased risk that future infections will be due to an antimicrobial resistant bacterium. Antibiotic treatment significantly increases the risk that subsequent infection is due to an antibiotic resistant bacterium.
A meta-analysis of four studies has shown that antibiotic treatment for a urinary tract infection results in a 2.5 times greater risk that a subsequent urinary tract infection in the next three months is due to an antibiotic resistant E. coli. Similarly, antibiotic treatment results in an overall 2.4 times greater risk that a respiratory tract infection in the subsequent 12 months is due to an antibiotic resistant S. pneumoniae, Haemophilus influenzae or S. pyogenes.
Antibiotic treatment results in a 3 times greater risk that any staphylococcal disease in the next three months is due to MRSA. 24 Patients deserve to be informed of these adverse effects, especially when seeking advice about a condition in which antimicrobial treatment makes no significant impact on mortality or morbidity.
Other risks of antibiotic use may be less obvious. Even brief courses of antibiotic treatment result in prolonged disruption of the normal flora of the gastrointestinal tract. 25 Such antibiotic caused alterations in the normal intestinal flora may then result in alterations in the person’s metabolic pathways with an increased risk of diabetes and obesity. 26,27
Patients need to be educated that interactions with the bacteria present in their mouth and intestines, and on their skin represent their most intimate relationship with the environment. They should be encouraged to appreciate and protect their normal flora and avoid unnecessary use of antibiotics that disrupt it.
New Zealand’s Ministry of Health should lead efforts to reduce antimicrobial consumption
The New Zealand Ministry of Health needs to reinvigorate efforts to reduce antimicrobial consumption in New Zealand by setting targets for reductions in antimicrobial consumption and then reporting on progress towards these targets by each DHB.
New Zealand is fortunate in already having excellent data on antibiotic consumption. Regular reporting of this data in relation to targets will encourage health administrators throughout the country to allocate resources to local or national programs that encourage reductions in antibiotic use.
Residents of DHBs that consistently fail to meet targets should be encouraged to press their health administrators for the reasons that they are failing to take the steps necessary to ensure that antibiotic treatments will remain beneficial for their population.Failure to act now to reduce antibiotic consumption will significantly harm the health of our population in the near future.