Canadian Antimicrobial Resistance Surveillance System Report 2016

Glossary

AMR

Antimicrobial resistance

AMU

Antimicrobial use

AST

Antimicrobial susceptibility testing

BCCDC

British Columbia Centre for Disease Control

BSI

Bloodstream infection

CA

Community-associated

CAHI

Canadian Animal Health Institute

CA-MRSA

Community-associated methicillin-resistant Staphylococcus aureus

CARA

Canadian Antimicrobial Resistance Alliance

CARSS

Canadian Antimicrobial Resistance Surveillance System

CDI

Clostridium difficile infection

CIDSC

Communicable and Infectious Disease Steering Committee

CIPARS

Canadian Integrated Program for Antimicrobial Resistance Surveillance

CNDSS

Canadian Notifiable Disease Surveillance System

CNISP

Canadian Nosocomial Infection Surveillance Program

CPHLN

Canadian Public Health Laboratory Network

CPO

Carbapenemase-producing organisms

CRA

Carbapenem-resistant Acinetobacter

CRE

Carbapenem-resistant Enterobacteriaceae

CTBLSS

Canadian Tuberculosis Laboratory Surveillance System

DDD

Defined daily doses

DID

Daily doses per 1000 inhabitants per day

EARS-NET

European Antimicrobial Resistance Surveillance Network

EIP

Emerging Infections Program

ESAC-Net

European Surveillance of Antimicrobial Consumption Network

ESAG

Enhanced Surveillance of Antimicrobial Resistant Gonorrhea

ESBL

Extended-spectrum ß-lactamase

ESVAC

European Surveillance of Veterinary Antimicrobial Consumption

GAS

Group A Streptococcus

GNB

Gram-negative Bacilli

HA

Healthcare-associated

HA-CDI

Healthcare-associated Clostridium difficile infection

HA-MRSA

Healthcare-associated methicillin-resistant Staphylococcus aureus

HO-CDI

Hospital-onset Clostridium difficile infection

HUS

Hemolytic Uremic Syndrome

ICU

Intensive Care Unit

iGAS

Invasive Group A streptococcal diseases

INH

Isoniazid

IPD

Invasive pneumococcal disease

IV

Intravenous

KPC

Klebsiella pneumoniae carbapenemase

MDR

Multidrug-resistant

MDR-TB

Multidrug-resistant Tuberculosis

MIC

Minimum inhibitory concentration

MRSA

Methicillin-resistant Staphylococcus aureus

MSM

Men who have sex with men

NARMS

National Antimicrobial Resistance Monitoring System for Enteric Bacteria

NHS

National Health Service

NHSN

National Healthcare Safety Network

NIHB

Non-Insured Health Benefits

NML

National Microbiology Laboratory

PCR

Polymerase chain reaction

PCU

Population Correction Unit

PCV13

13-valent pneumococcal conjugate vaccine

PFGE

Pulsed-field gel electrophoresis

PHAC

Public Health Agency of Canada

PHN

Public Health Network

RMP

Rifampin

SSTI

Skin and soft tissue infection

STI

Sexually transmitted infection

TB

Tuberculosis

TMP-SMX

Trimethoprim-sulfamethoxazole

US CDC

United States Centers for Disease Control and Prevention

UTI

Urinary tract infections

VRE

Vancomycin-resistant Enterococci

VRSA

Vancomycin-resistant Staphylococcus aureus

WHO

World Health Organization

XDR-TB

Extensively drug-resistant TB

Message from the Chief Public Health Officer and the President of the Public Health Agency of Canada

Antimicrobial resistance (AMR) continues to be a serious public health issue in Canada and internationally. Common and treatable infections may once again become deadly. This reality is illustrated by the recent detection of the mcr-1 colistin resistant gene in isolates from animal, food, and human sources in Canada, the United States and elsewhere in the world.

Last year, the inaugural “Canadian Antimicrobial Resistance Surveillance System (CARSS) Report 2015” presented integrated surveillance data from nine of the Public Health Agency of Canada (PHAC)’s existing surveillance systems and laboratory reference services^{Footnote 1} to provide information on AMR and AMU in Canada, with the aim to support decision-making by health professionals and policy makers.

This second report reflects a year of momentum, building on the first report by increasing the depth and breadth of surveillance data and analysis.

Highlights of this report include the identification of priority organisms to help focus surveillance efforts; results of a pilot of enhanced surveillance of antimicrobial-resistant gonorrhea to better support the response to the resurgence of gonorrhea reported in recent years; and outcomes of a feasibility study of “AMR-Net”, a web-based application that was used to collect and analyze community-level AMR information and which could serve to educate healthcare professionals and Canadians about AMR trends in the community. The report also identifies surveillance gaps which future CARSS reports will strive to fill to enable us to provide a more comprehensive picture of AMR and AMU in Canada.

PHAC’s surveillance efforts can only be successful with the contributions and collaboration of its partners, the provinces and territories, healthcare professionals, veterinarians, associations and organizations, communities and individuals. These partners provide data and advice essential to making CARSS the national focal point for AMR and AMU surveillance data in Canada. We thank all contributors for their time, expertise and continued support.

Dr. Gregory Taylor
Chief Public Health Officer of Canada

Dr. Siddika Mithani
President, Public Health Agency of Canada

Executive Summary

Introduction

The establishment of CARSS is a key commitment made in the Federal Action Plan on AMR and AMU in Canada: Building on the Federal Framework for Action. CARSS provides an integrated picture of AMR/AMU in Canada based on available surveillance data from PHAC’s nine surveillance systems and laboratory reference services which track the identified priority organisms. The inaugural CARSS report issued in 2015 provided information on AMR/AMU in Canada until 2013. This year’s report demonstrates the Government of Canada’s continued commitment to leading activities to prevent, limit and control the emergence and spread of AMR as described in Antimicrobial Resistance and Use in Canada: A Federal Framework for Action.

One of the key accomplishments this year has been the work with federal/provincial/territorial partners to develop a list of priority organisms of concern for AMR. The list was developed by leveraging the expertise of the federal/provincial/territorial Pan-Canadian Public Health Network’s (PHN’s) Communicable and Infectious Disease Steering Committee (CIDSC). Working with PHAC, a task group identified and prioritized AMR organisms of concern in Canada; identified where sufficient surveillance data are being collected and where gaps may exist; and made recommendations to address some of those gaps. The list of priority organisms is an important step in determining the data that need to be collected from all surveillance partners in Canada.

Since the release of the CARSS 2015 report, PHAC has invested significant effort in AMR/AMU surveillance including the implementation of two pilot initiatives to address gaps in community settings identified in the initial CARSS report and the spring 2015 Auditor General of Canada report “Antimicrobial Resistance”. The Enhanced Surveillance of Antimicrobial Resistant Gonorrhea (ESAG) pilot was undertaken in response to the high levels of resistance to antimicrobials used for treating gonorrhea. ESAG assessed the feasibility of obtaining surveillance data to improve the understanding of the current levels and trends of resistant gonorrhea in Canada. It aimed to provide stronger evidence to guide the development of treatment guidelines and public health interventions to minimize the spread of antimicrobial resistant gonorrhea. The AMR-Net pilot was undertaken to determine the feasibility of obtaining and analyzing existing antimicrobial susceptibility data in the community; to improve PHAC’s ability to respond to emerging threats and support stewardship efforts by informing evidence-based decision making. The results of the two pilot initiatives confirmed the feasibility of collecting data in community settings. 

Antimicrobial resistance and antimicrobial use

Antimicrobial resistance is the ability of microorganisms (including bacteria, fungi, viruses and parasites) to become resistant to treatment by antimicrobial drugs, such as antibiotics^{Footnote 2}. Resistance can develop naturally over time as microorganisms evolve, mutate and multiply. Microorganisms, especially bacteria, are also able to transfer their resistant traits to other microorganisms, increasing the spread of AMR. The scope of resistance is accelerated by excessive and/or unnecessary use of antimicrobial drugs used to treat bacterial infections, such as inappropriate prescribing practices by health professionals and patients not taking prescribed drugs as directed.

Antimicrobial resistance

Antimicrobial resistance significantly impedes our ability to fight infectious diseases, leading to more hospitalizations and/or increased hospital stays. This results in increased health care costs and costs to society in the form of time away from work, increased disability claims and loss of productivity.

Antimicrobial resistance is now occurring in every reporting country in the world. Globally, bacteria such as Staphylococcus aureus, Escherichia coli, and Klebsiella pneumoniae, are demonstrating a reported range of resistance of between 5% and 80% of tested strains. Current surveillance in Canada is showing generally stable rates of resistance and in some cases a decline of infection rates of select AMR organisms in recent years. For example, rates of hospitalized methicillin-resistant Staphylococcus aureus (MRSA) infection have declined 25% since 2008, and declines in each of the past two years for vancomycin-resistant Enterococci (VRE) infections have been observed. That said, incidence rates of MRSA and VRE have not decreased to levels seen before 2007 when increases in resistance began, indicating that more work needs to be done to reverse the problem.

Although Canada is not seeing the same general level of resistance as some other countries, there are areas of concern. For example, available data indicate that more than one third of gonorrhea cases are resistant to each of ciprofloxacin, erythromycin and tetracycline. In recent years, there has been a small proportion of cases (<0.3%) resistant to both azithromycin and cephalosporins (ceftriaxone and cefixime) the currently recommended dual therapy treatment^{Footnote 3} for gonorrhea. Canada demonstrates higher azithromycin resistance levels to N. gonorrhoeae (3.3% in 2014) than the United States which reported 0.6% in 2013 and the United Kingdom which reported 1.6% in 2013^{Footnote 4},^{Footnote 5}.

If the rate of resistance continues to increase, it would threaten the success of the currently recommended dual therapy treatment regimen. This shows the need for ongoing monitoring of AMR to help maintain the effectiveness of current treatment regimens and to guide their modification when appropriate.

Antimicrobial use in humans

In Canada, the majority of antimicrobials used by humans are available by prescription only. Over the last 13 years, the volume of antimicrobial prescriptions in humans has remained relatively stable. In 2014, 23 million prescriptions were dispensed, with 93% dispensed by community pharmacies. The overall expenditure on antimicrobials in Canada was $786M, with community dispensing accounting for 87% and hospital purchases accounting for 13% of this amount.

Antimicrobials were most often recommended for treating respiratory infections. Eighty-two percent (82%) of acute sinusitis diagnoses, 77% of acute bronchitis diagnoses, and 74% of pneumonia diagnoses resulted in a recommendation for an antimicrobial. Further work is required to assess the appropriateness of practitioner antimicrobial recommendations and adherence to clinical guidelines for specific antimicrobials for first-line treatments. An example would be the treatment of lower urinary tract infections where ciprofloxacin is recommended more often than the guidelines suggested trimethoprim-sulfamethoxazole or nitrofurantoin. In 2014, a total of 30 European countries provided information to the European Surveillance of Antimicrobial Consumption Network (ESAC-Net) on antimicrobials consumed in their community. When these data were compared with the 2014 Canadian outpatient antimicrobial consumption rate, Canada (17.8 defined daily dosage (DDD)s per 1,000 persons per day) ranked 12th out of 31 countries by increasing level of AMU, with almost half the level of use reported by Greece (country with highest use, 34 DDDs per 1,000 persons per day)^{Footnote 6}.

There is significant variation in the rates of antimicrobial prescribing within Canada. Differences observed in prescriptions filled for parenteral (injection and intravenous) antimicrobials through community pharmacies likely reflect differences in provincial policies regarding the payment for outpatient parenteral antimicrobials. In contrast to the large variation observed among the top five antimicrobials purchased by hospitals, commonalities are observed among the antimicrobials dispensed in the community across provinces and territories. For example, amoxicillin and clarithromycin were among the top five antimicrobials dispensed with the highest levels of defined daily doses (DDDs) per inhabitant in every province and territory. Further analysis is required to determine the reason for the observed differences.

Antimicrobial use in animals

There is increasing evidence that the use of antimicrobial agents in veterinary medicine and livestock production is an important contributing factor to the emergence and persistence of AMR in bacteria in humans. The spread of organisms with resistance traits from animals to humans necessitates the assessment of human health risks associated with AMU in food-producing animals.

Canada is a major source of food-producing animals for domestic and international markets, with approximately 19 times the number of animals than humans in the country. The majority (73%) of antimicrobials distributed to animals were in the same classes as those antimicrobials used in human medicine. In 2014, approximately 82%^{Footnote 8} of antimicrobials important to human medicine were distributed and/or sold for use in food-producing animals. Antimicrobials are used in food-producing animals (e.g., chickens, pigs and cattle) for the treatment and prevention of disease, and to improve feed efficiency/promote growth.

In Canada, as in many other countries, AMR is monitored in chicken, pork and beef for Escherichia coli, Campylobacter and Salmonella as a way to measure the potential movement of AMR priority organisms from animals to humans. In comparison to the countries participating in the European Surveillance of Veterinary Antimicrobial Consumption (ESVAC) network, Canada ranked 7th highest out of 27 countries for increasing levels of antimicrobial sales adjusted by populations and weights^{Footnote 9}. Canada’s total milligrams distributed, adjusted by population, was 44 times that used in Norway (country with the lowest sales) and less than half of that reported by Cyprus (the country with the highest sales).

AMR priority organisms

PHAC’s surveillance systems are essential tools in its efforts to protect Canadians from the health risks associated with AMR. Surveillance data inform the understanding of transmission routes and help assess the magnitude and trends of related morbidity and mortality. In this context, PHAC identified 138 infectious pathogens (microorganism that can cause disease) worldwide that have exhibited resistance, and from these used a multi-step approach to assess and determine the AMR pathogens of relevance to Canada. The criteria used included the incidence, communicability/transmissibility, preventability, treatability, clinical impact and mortality.

Subsequently the PHN’s CIDSC established a task group to determine the pathogens of greatest importance to public health in Canada. Led by PHAC, it included epidemiologists, infectious diseases and public health experts from across the country, and federal/provincial/territorial experts.

Although it was recognized that resistance can occur in all types of organisms including viruses, fungi and parasites, the task group focused on bacteria in humans and animals as it relates to human health. The group identified the following organisms as the first order of priority for surveillance purposes, which should be, or currently are, under surveillance in Canada:

• Clostridium difficile
• Extended-spectrum β-lactamase (ESBL) -producing organisms^{Footnote 10}
• Carbapenem-resistant organisms^{Footnote 11} (Acinetobacter + Enterobacteriaceae spp.)
• Enterococcus spp.
• Neisseria gonorrhoeae
• Streptococcus pyogenes (Group A Streptococcus) and pneumoniae
• Salmonella spp.
• Staphylococcus aureus
• Mycobacterium tuberculosis
• Campylobacter spp.

Organisms transmitted in healthcare settings

PHAC’s surveillance systems currently monitor the following priority bacteria transmitted mainly in healthcare settings:

Clostridium difficile (C. difficile) is an important healthcare-associated infection that causes significant morbidity and mortality. While largely responsive to current standard treatments, the infection spreads rapidly because it is naturally resistant to many drugs used to treat other infections. Most cases occur in patients taking high doses of specific antibiotics or taking antibiotics over long periods of time, those who have underlying medical conditions, or those who have had invasive medical procedures. Following a decade of increasing rates of C. difficile in hospitalized patients, rates declined in 2014 from 5.2 cases per 1,000 patient admissions to 3.4 cases per 1,000 patient admissions. 

Carbapenem resistant Organisms (CRO) and Carbapenem-resistant Enterobacteriaceae (CRE) can cause infections in the urinary tract, respiratory system, bloodstream or in wound infections of vulnerable patients (such as the very young, elderly or immune-compromised). The carbapenem group of antimicrobials is most often a last line of treatment for Enterobacteriaceae and is usually reserved for multidrug-resistant Enterobacteriaceae.  The emergence of CRE and a related group called Carbapenemase-producing organisms (CPO) have become serious public health concerns internationally because there are few alternative treatments available when resistance to this group of antimicrobials occurs. In Canada, both the CRE and CPO rates have remained consistently low since 2010 but continue to be closely monitored given the increasing rates in other countries.

Methicillin-resistant Staphylococcus aureus (MRSA) is one of the most common causes of healthcare-associated infections in Canada. Methicillin is one of the first-line antibiotics used to treat Staphylococcus aureus infections.  MRSA can cause serious and sometimes fatal infections in the hospital setting. With resistance to methicillin, vancomycin has now become the primary antimicrobial used to treat MRSA infections. As a result, resistance to vancomycin needs to be monitored to inform action to preserve this treatment option. Since 2008, the overall MRSA infection rates in hospitals have decreased by 25%, but have not declined to rates seen in the early 2000s.

Vancomycin-resistant Enterococci (VRE) can cause a range of infections in the urinary tract, bloodstream and surgical wounds of hospital patients. Vancomycin is generally prescribed to treat serious infections caused by organisms that are resistant to other antimicrobials. As such, resistance to vancomycin further limits therapeutic options. VRE infection rates increased sharply from 2007 (0.10 cases per 10,000 patient days) to 2012 (0.61 cases per 10,000 patient days) before decreasing in 2013 and 2014 (to 0.45 cases per 10,000 patient days). Despite recent decreases in Canada, VRE infections are increasing in prevalence worldwide and therefore, continue to be a global health threat that could arise again in Canada. 

Organisms transmitted in community settings

The following priority bacterial organisms transmitted primarily in community settings are being monitored through PHAC’s surveillance systems:

Streptococcus pneumoniae (S. pneumoniae) causes a severe form of infection that can lead to pneumonia and meningitis most often diagnosed in young children and the elderly. S. pneumoniae has shown resistance to penicillin and the erythromycin group of drugs. The national annual incidence rate of invasive pneumococcal disease and the resistance to a number of antimicrobials used to treat S. pneumoniae have decreased since 2010 with the implementation of childhood immunization programs using a 13-valent pneumococcal conjugate vaccine.

Group A Streptococcus (GAS) can cause invasive diseases such as necrotizing fasciitis (flesh-eating disease) and non-invasive diseases such as strep throat and scarlet fever. From 2009 to 2013, the national incidence rate of invasive GAS cases increased from 4.0 to 4.7 per 100,000 population, with the highest incidence seen in infants less than one year of age and in seniors  60 years of age and older. In 2014, all samples of invasive GAS responded to first-line antimicrobials, while resistance to second-line antimicrobials either remained relatively unchanged or declined.  The serious nature of the infections caused by GAS requires that AMR be closely monitored to ensure effective treatments remain available.

Neisseria gonorrhoeae (gonorrhea) can cause genital/reproductive tract inflammation and damage, and potentially, infertility. Between 2004 and 2013, the rate of reported cases of gonorrhea increased 43.1% (from 27.4 to 39.3 per 100,000 population), particularly in females. Coupled with the increase in cases, N. gonorrhoeae has developed resistance to a range of antimicrobials used to treat it. In 2014, 18.2% of isolates were resistant to penicillin; 47.3% were resistant to tetracycline; 32.0% were resistant to erythromycin; and 34.0% were resistant to ciprofloxacin. There has also been decreased susceptibility to cefixime, ceftriaxone or azithromycin, threatening the availability of treatment options.

Mycobacterium tuberculosis (TB) rates in Canada are among the lowest in the world; however, it disproportionately affects Indigenous people and the foreign-born individuals from areas of the world with high rates of TB. Overall, there was no notable change in the resistance to first-line medications in Canada from 2004 to 2014 and resistance levels remain low (8% resistance to one drug and 1% multidrug-resistance) and below international levels. Although AMR TB is not a major problem in Canada, the potential for the emergence of more cases in Canada exists due to increasing levels in other countries and ease of international travel. 

Organisms transmitted in animals

The following priority bacteria transmitted through animals are monitored through PHAC’s Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) system:

Escherichia coli (E. coli) are extremely common bacteria, as they inhabit the gastrointestinal tract of healthy animals and humans. They are considered to be good indicators of antimicrobial selection pressure, can carry resistance genes that can be spread to other bacteria, and, in some cases, may cause illness themselves. In 2014, the proportion of chicken, swine and cattle samples (on farm, at slaughter and at retail) positive for generic E. coli were 96%, 55% and 56% respectively. Following the May 2014 industry ban of the preventive use of antimicrobials considered of very high importance to human medicine, including ceftiofur, a third generation cephalosporin, CIPARS observed a decrease in resistance to third generation cephalosporins in E. coli from chickens and chicken meat. For example, 28% of isolates from chicken meat were resistant to third generation cephalosporins in 2013 but only 19% in 2014. PHAC also observed a decrease in the use of ceftiofur in broiler chicken flocks from 31% in 2013 to 6% in 2014.

Campylobacter are a major cause of foodborne diarrhoeal illness in humans. The majority of Campylobacter infections are mild, but can be severe or even fatal among very young children, the elderly and immune-compromised individuals. In 2014, the proportion of chicken, swine and cattle samples (on farm, at slaughter and at retail) positive for Campylobacter were 25%, 73% and 87% respectively. CIPARS observed a decrease, corresponding with the industry ban on the preventive use of antimicrobials of very high importance to human health, in the number of farms reporting fluoroquinolone use from 2/99 in 2013 to 0/142 in 2014.

Salmonella spp .Salmonellosis is one of the most common and widely distributed foodborne diseases, with tens of millions of human cases occurring worldwide every year. Caused by over 2,500 different types of Salmonella, the majority of infections cause mild gastroenteritis but can be severe in the young, the elderly and the immune-compromised. In 2014, the proportion of chicken and swine samples (on-farm, at slaughter and at retail) positive for Salmonella spp. were 34% and 16% respectively. As in E. coli there was a decrease in resistance to third generation cephalosporins in non-typhoidal Salmonella from chickens and chicken meat in 2014 compared to 2013. For example, 22% of isolates from chickens sampled on-farm were resistant to third generation cephalosporins in 2013 but only 12% in 2014. By comparison, resistance to third generation cephalosporins in Salmonella from pigs was below 5% in 2014.

Surveillance data gaps

PHAC in collaboration with provinces and territories has identified gaps in surveillance which require addressing in order to provide a comprehensive picture of AMR/AMU in Canada. Of most importance is the lack of data regarding the resistance of a number of priority pathogens of concern to the health of Canadians. This includes resistant E. coli, Neisseria gonorrhoeae, and Shigella. Other gaps include limited data on AMR in smaller, non-academic hospitals, no data for rural and northern healthcare settings and First Nations and Inuit communities, and limited data on AMR in the community, outpatient clinics, long-term care facilities, and physicians’ and dentists’ offices.

As resistant microbes are generally found where AMU is more prevalent, identifying and addressing gaps in AMU data is critical to mitigating the threat of AMR. PHAC obtains information on prescriptions dispensed by retail pharmacies, antimicrobials purchased by hospitals, and diagnoses for which physicians have recommended an antimicrobial in the community^{Footnote 12}.  While there is some understanding of prescribing patterns in hospital and community settings, other information gaps limit the ability to assess appropriate indications of use and identify potential areas of overprescribing.

Currently, there is limited data on AMU in animals as well as some gaps in the surveillance of AMR in animals and the food chain. There is a lack of information on the quantities of antimicrobials imported for own use or as active pharmaceutical ingredients for compounding by veterinarians and used in food-producing animals. The introduction of new federal regulations for veterinary oversight of AMU is anticipated to open new sources for some of these data after regulations are passed in December 2016.

Finally, there is a significant gap in understanding the linkages between AMU and the observed patterns of resistance and the spread of pathogens in Canada.

Next steps/conclusion

PHAC has made significant progress strengthening its surveillance systems in order to provide a comprehensive and integrated public health picture of AMR and AMU in Canada. Ongoing surveillance gaps present a challenge to developing a comprehensive picture in both the community and hospital settings. PHAC is committed to working with provincial and territorial governments, and other partners to address surveillance gaps. The identification of the increasing resistance of microorganisms to antimicrobials, and questions regarding the appropriate use of antimicrobials, demonstrate the necessity for continued vigilance in order to address the issue of AMR and AMU in Canada. Expanding surveillance activities to collect quality data regarding health professional prescribing practices, infection rates, and resistance patterns for priority organisms, particularly in the community setting, will be the priority of PHAC’s work over the coming year. 

AMR is a global problem that cannot be solved without global collaboration. PHAC will continue to work with its international partners such as the World Health Organization (WHO) to identify and measure common indicators as well as share best practices for stewardship to tackle AMR in Canada.

Technical annex

Antimicrobial resistance and antimicrobial use

The following technical annex provides a more detailed view of the surveillance data on the priority organisms surveilled under PHAC’s surveillance systems. The AMR section focuses on priority organism and describes the surveillance methods, the situation in Canada and an international perspective. The priority organisms listed in the following AMR section includes: Clostridium difficile, multidrug-resistant Enterobacteriaceae and Acinetobacter; methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus spp.; drug-resistant Streptococcus, Neisseria gonorrhoeae, Salmonella typhi, non-Typhoidal Salmonella, and Tuberculosis.   

The AMU section provides information on human and animal antimicrobial usage. It gives details on the human aspect for the amount of antimicrobials dispensed through community pharmacies, prescriber specialization breakdown as well as hospital purchasing and use. The animal aspect is presented by antimicrobials distributed for sale, indication for use and international comparison.

Antimicrobial resistance in Canada

Clostridium difficile

Clostridium difficile (C. difficile) is a gram-positive, spore-forming, anaerobic bacillus and is the most frequent cause of healthcare-associated infectious diarrhea in Canada and other developed countries. There has been a marked increase in C. difficile infection (CDI) incidence and mortality across the United States, Canada and Europe during the last decade^{Footnote a}. Most cases of CDI occur in patients who are elderly and who have other underlying medical conditions. C. difficile spreads rapidly in healthcare settings by direct contact because it is naturally resistant to many antimicrobials used to treat other infections. C. difficile spores in the environment also tend to be resistant to most commonly-used disinfectants. It is estimated that 1.5% of all hospitalized patients will develop CDI during the course of their hospitalization. Approximately 15% of these will develop severe disease and 6% of severe CDI cases will die from the infection^{Footnote b,Footnote c}.

While resistance to current standard treatments for CDI(vancomycin, metronidazole)is not a concern at present, the extensive use of antimicrobials in hospitals to treat patients often creates a competitive advantage for C. difficile. A virulent strain of C. difficile, the North American pulsed-field (NAP) type 1 (NAP1) which emerged in 2000 was found to be resistant to fluoroquinolones, a class of antimicrobials commonly used to treat other infections in both the hospital and community settings. NAP1 has spread throughout North America and Europe and is responsible for increasing CDI rates, hospital outbreaks and more severe disease especially in the elderly^{Footnote c}.   Whereas NAP1 appears to be most prevalent in adult cases of CDI, NAP4 was found to be the most common type found in pediatric cases with CDI between one and 18 years of age^{Footnote d}.

Methods

PHAC has collected information on healthcare-associated (HA)-CDI through the Canadian Nosocomial Infection Surveillance Program (CNISP) since 2005. As of 2014, CNISP collected information in 58 major hospitals in 10 provinces.  Infection Control Professionals complete a standardized patient questionnaire through concurrent or retrospective chart review once a possible CDI case is identified by the hospital’s laboratory. Detailed patient information including outcome information (admitted to intensive care unit for complications related to HA-CDI, underwent colectomy or died) was submitted to PHAC for each identified case of HA-CDI.  Attributable mortality was provided from March 1st to April 30th each year. Whenever possible, frozen stool specimens from patients with laboratory confirmed HA-CDI were forwarded to the National Microbiology Laboratory (NML) for C. difficile isolation and molecular characterization. Stool specimens in adult patients 18 years of age and older with HA-CDI were submitted between March 1st and April 30th of each year whereas the  stool specimens of children aged one year to less than 18 years were submitted year round to NML.

CDI in Canada

In 2008, the infection rate peaked at 5.8 cases per 1,000 patient admissions and 7.5 cases per 10,000 patient days. Since 2011, the rates have been slowly declining from 5.2 cases per 1,000 patient admissions to 3.4 in 2014. Similarly, since 2011, the infection rates per 10,000 patient days have fallen from 6.6 cases to 4.4 in 2014 (Figure 1).

In 2014, mean age for adults (n= 2,599) was 68.1 years (range: 18 years to 102 years) and 51% (n = 1,324) were males. The mean age for the pediatric cases (n= 133) was 7.5 years (range: 1 year to 17 years) and 56% (n = 74) were males.

Adults continue to have higher reported rates than pediatric age groups^{Footnote 13}. Since 2011 the rates for adults appear to be declining whereas the pediatric rates appear to be relatively more stable (Figure 2)^{Footnote 14}.

Data on outcomes, 30 days after the date of first positive C. difficile test, were available for 512 adult cases and 107 pediatric cases in 2014. Of the 512 adults, 11.9% (n = 61) were reported to have died. CDI was reported to have contributed to but was not the cause of death for 19 cases and CDI was the cause of death for 6 cases. The causal relationship between CDI and death was not known for 15 cases. For the remaining 21 cases, CDI was not related to the individual’s death. For the 107 pediatric cases, there was one death and CDI was not related to that outcome.

In 2014, the results of strain typing were reported for 388 adult cases and 97 pediatric cases. For the adult cases, 28.4% of C. difficile strains were the North American Pulsed Field Type 1 (NAP1) strain whereas only 4.1% of the pediatric cases were strain type NAP1 where 23.7% of the pediatric cases were NAP4 (Figure 3).

Antibiogram results were available for 2,290 adult cases and 518 children diagnosed between 2007 and 2014. Susceptibility results were reported for the following antimicrobial agents: clindamycin, metronidazole, moxifloxacin, rifampin, tigecycline and vancomycin. For adults, between 2007 and 2014, no resistance to metronidazole or tigecycline was reported and one case was found to be non-susceptible to vancomycin. In 2009, 63.3% of adult cases were found to be resistant to moxifloxacin. Between 2009 and 2014, the percentage of cases found to be resistant has been slowly declining and by 2014, 34.3% of the reported cases were resistant to moxifloxacin. In 2009, 37.5% of the reported cases were resistant to clindamycin and resistance level fell to a low for the period of 13.4% of reported cases by 2012. However, in 2013, 30.7% of cases were found to be resistant to clindamycin and this went up to 41.3% again in 2014.

Compared with adults, resistance to clindamycin was reported more frequently among children. In 2009, 62% of reported cases were resistant and although there was a dramatic decline observed in 2010 to 13.5%, from 2011 to 2014 the proportion of cases showing resistance increased to 51.5%. In children, the proportion of cases resistant to moxifloxacin has been lower than in adults. In 2009, 14.3% of children were resistant to moxifloxacin. Between 2010 and 2012 the rate remained relative stable but in 2013, 5.0% of reported cases were resistant to moxifloxacin. In 2014 the proportion increased slightly to 5.2%.

International perspective

The lack of internationally-recognised standardized surveillance case definitions for CDI across countries hampers the interpretation of CDI epidemiology. In the United States, C. difficile surveillance is performed through two systems: the Emerging Infections Program (EIP) which is an active, population- and laboratory-based surveillance system across diverse United States (USA) geographic locations. The EIP started in 2009 and currently runs in 10 EIP sites with approximately 11.7 million people under surveillance. On the other hand; hospital data are collected through the National Healthcare Safety Network (NHSN) since 2013^{Footnote e}.

Comparing Canadian and the United States CDI rates is somewhat difficult as the United States data are calculated with population estimates as opposed to the Canadian national rates that are reported in patient days. However, the declining rates reported in Canada were also seen in the United States. A 10% decrease in hospital-onset CDI (HO-CDI^{Footnote 15}) was reported between 2011 and 2013 in the United States^{Footnote f}.  In 2010, the overall United States HA-CDI^{Footnote 16} rates ranged from 58.5 to 94.8/100,000 population across seven EIP sites^{Footnote g}. In 2011, the average reported rate across 10 sites was 92.8 cases/100,000 population^{Footnote h}. Also, the NAP1 strain was more prevalent in the HA-CDI than among community-associated infections (30.7% vs. 18.8%, P<0.001) in 2011^{Footnote h}. Deaths related to C. difficile increased 400% between 2000 and 2007 in the United States. About half of CDI cases occur in people less than 65, yet more than 90% of deaths are reported in people ≥ 65^{Footnote i}.

In England and Wales, voluntary surveillance for CDI has been conducted since 1990. National Health Service (NHS) acute trust mandatory reporting, in people aged ≥65 years, was introduced in January 2004. Since 2007, mandatory NHS acute trust surveillance was enhanced to include all patients ≥2 years and for all C. difficile infections. As per Public Health England, the overall mandatory “trust apportioned CDI”^{Footnote 17} rates have decreased between 2007-2008 (89.6 per 100,000 bed days) and 2014-2015 (15.1 per 100, 000 bed days). However, there has been a 2.9% increase between 2013-2014 (14.7 per 100,000 bed days) and 2014-2015. As seen in Canada and USA, there is a notable declining trend in the proportion of reported HO-CDI (≥3 days post-admission) from 78.8% in 2007-2008 to 57.4% in 2014-2015 in the United Kingdom^{Footnote j}.

Multidrug-resistant Enterobacteriaceae and Acinetobacter

Enterobacteriaceae are gram-negative bacilli (GNB) commonly encountered in both the healthcare settings and in the community; and include species such as Escherichia coli, Klebsiella species (spp.) and Enterobacter spp. These organisms are normal flora in the gastrointestinal tract in healthy humans (referred to as colonization), but can cause infections in vulnerable individuals, including urinary tract, respiratory, bloodstream, and wound infections. Enterobacteriaceae have mixed susceptibility to commonly-prescribed antimicrobials. One form of resistance occurs when GNB acquire the ability to produce extended-spectrum β-lactamases (ESBL), an enzyme class that renders it resistant to commonly-used extended-spectrum (third-generation) cephalosporins (e.g., ceftazidime, cefotaxime and ceftriaxone) as well as beta-lactam-lactamase inhibitor combinations (e.g., piperacillin-tazobactam, etc.)^{Footnote k}.

Another form of resistance involves the carbapenem group of antimicrobials which is a relatively safe and generally effective treatment for ESBL as well as other highly resistant gram-negative organisms and there are few alternative treatments available when resistance to carbapenem occurs. Enterobacteriaceae that have acquired resistance to carbapenems are called carbapenem-resistant Enterobacteriaceae (CRE). Some CRE are also carbapenemase-producing organisms (CPO) by virtue of their ability to produce enzymes, called carbapenemases, which break down carbapenems^{Footnote k}. There are other GNB outside of the Enterobacteriaceae family that have also shown resistance to carbapenems including Acinetobacter spp. and Pseudomonas spp.

Methods

CNISP has been collecting data on CRE and CPO in hospitalized patients since 2010^{Footnote l}. Participation in this surveillance has increased from 33 CNISP hospitals in 2010 to 58 CNISP hospitals in 2014.  All CRE and CRA identified in the participating hospitals are submitted to NML for further testing. If an isolate was determined to be a carbapenemase producer by NML, a detailed patient questionnaire was completed.  Detailed patient information included patient demographics, where the culture was obtained (urine, wound, etc.), travel history and whether or not the patient had died. 

Carbapenem-resistant Enterobacteriaceae in Canada

Between 2010 and 2014, there were a total of 652 CRE cases collected from the CNISP sites. The CRE rate did not change significantly from 2010 (0.19 per 10,000 patient days) to 2014 (0.22 per 10,000 patient days); however, there were two high peaks in 2011 and 2013, in which the incidence rate nearly doubled from that of 2010 to reach an incidence rate of 0.32 per 10, 000 patient days in 2011 and 0.28 per 10,000 patient days in 2013 (Figure 4). The increase in both years was largely due to an ongoing outbreak at one individual hospital and was not reflective of a national trend.

Carbapenemase-producing organisms in Canada

Between 2010 and 2014, CNISP hospitals reported a total of 301 CPO isolates, representing 273 individuals. Among those individuals, 25 (9.1%) harbored at least two CPOs. Nationally, the incidence of CPO has remained stable from 2010 (0.10 per 10,000 patient days) to 2014 (0.11 per 10,000 patient days) (Figure 5).

Figure 6 shows the distribution of reported CPO cases by gender and age over the 2010 to 2014 period for 269 individuals. Overall, males represented 62% of all cases (n=167) and individuals 65 and over (n=158) represented 58% of all CPO cases.

Among the 273 individuals harboring CPO’s from 2010 to 2014, 105 (38.8%) were infections, 131 (48%) were in patients who harbored the bacteria without showing signs of infection (colonization) and the status was unknown for 38 (13.9%) cases. The majority of the reported CPO isolates was captured from rectal/stool swabs (n=91). Rectal and stool specimens represented 46% of all reported CPO cases which represents 22% of CPO infections and 69.5% of CPO-colonized cases. The highest number of CPO infections (n=106) were isolated in urine (n=72, 67%), followed by blood (n=24, 23%) then sputum, skin, stool and surgical sites (Figure 7).

From 2010 to 2014 the proportion of patients who died at 30 days after the date of first positive CPO culture has remained relatively stable ranging from 15% in 2010 to 12% in 2014. Death is reported as ‘all-cause mortality’ and is not necessarily attributable to the CPO infection. Treatment data were available for 2010 to 2013 for 122 out of the 206 individuals harboring CPOs (59.2%). The most commonly-prescribed antimicrobials from 2010 to 2013 were B-lactams (32.0%) followed by glycopeptides (29.5%), fluoroquinolones (22.0%) and carbapenems (19.7%). In 2013, the use of colistin and tetracycline to treat CPO infections increased; whereas, the use of B-lactams and glycopeptides decreased from 2010 to 2013 (Figure 8). Collection of antimicrobial treatment data was discontinued in 2014.

A total of 303 isolates that tested positive for a carbapenemase gene (n=303) by both polymerase chain reaction (PCR) and Etest at the National Microbiology Laboratory (NML) were summarized in Figure 9 based on the corresponding type of Enterobactericeae and Acinetobacter species.  Certain carbapenemases are specific to certain genera. SME are exclusive to Serratia spp., to be more specific Serratia marcescens and OXA-51, OXA-23 unique to Acinetobacter baumannii and NMC to Enterobacter cloacae. This was further supported by the results of the lab data collected from 2010 to 2014, where SME was only identified in Serratia spp.OXA51-OXA 23 in Acinetobacter spp.and NMC in Enterobacter spp. Klebsiella-pneumoniae carbapenemase (KPC) carbapenemase-producing organisms expands to several genera such as Klebsiella, Enterobacter, Citrobacter, Serratia and E. coli.^{Footnote m,Footnote n}. It is dominant in Klebsiella species which is expected as it was originally assumed to be unique to K. pneumoniae. The GES enzymes are dominant in Pseudomonas aeruginosa; however, they have been observed in many other genera of the Enterobacteriaceae family including Klebsiella, Citrobacter, Serratia spp.and E. coli^{Footnote m,Footnote n,Footnote o}. In these isolates, the New Delhi metallo-beta-lactamase (NDM) and oxacillinase (OXA)-48 enzymes are most commonly found in Klebsiella pneumoniae and E. coli; however, since this gene can easily spread from one microorganism to another by horizontal gene transfer, it is also seen in other species^{Footnote p}.

Carbapenemase-producing Enterobacteriaceae (CPE)

Between 2010 and 2014, CNISP reported on 270 isolates harboring CPE. The overall CPE rates have increased by approximately 33% from 2010 (0.06 per 1,000 patient admissions) to 2014 (0.08 per 1,000 patient admissions). However there was a high peak in 2011, in which the incidence rate doubled compared to 2010 (0.12 per 1,000 patient admissions) (Figure 10).

Carbapenemase-producing Acinetobacter (CPA)

Between 2010 and 2014, CNISP reported on 31 isolates harboring CPA. Following a peak in CPA incidence rates in 2010 (0.02 per 1,000 patient admissions, 0.03 per 10,000 patient days), the overall CPA rates per 1,000 patient admissions remained stable since 2012 (0.01 per 1,000 patient admissions) (Figure 11).

International perspective

Carbapenem resistance is under regular surveillance by the European Antimicrobial Resistance Surveillance Network (EARS-NET). AMR data are expressed as a percentage of resistant isolates out of all isolates with susceptibility information on carbapenem resistance for a specific pathogen. In Europe, between 2011 and 2014, carbapenem resistance significantly increased for K. pneumoniae (6.0% in 2011 to 7.3% in 2014), yet it remained relatively stable at 0.1% for E. coli. From 2011 until 2014, significantly increasing resistance trends to carbapenems were observed in seven countries (Bulgaria, Croatia, France, Germany, Italy, Portugal and Spain) for K. pneumoniae. Yet, in 2014, national resistance to carbapenem was 0.0 % in Estonia, Finland, Iceland, Norway and Sweden for K. pneumoniae. Also, no carbapenem resistance for E. coli was observed in Croatia, Cyprus, the Czech Republic, Estonia, Finland, Hungary, Iceland, Latvia, Lithuania, Malta, the Netherlands, Norway, Slovakia, Slovenia or Sweden. Resistance to a combination of antimicrobials including carbapenem remained low in E. coli (less than 0.1%) in 2014. On the other hand, approximately 5.7% of all K. pneumoniae isolates were resistant to the combination of fluoroquinolone, third-generation cephalosporins, aminoglycosides and carbapenem^{Footnote q}.

In 2015, the United States Centers for Disease Control and Prevention (US CDC) changed the CRE surveillance definition to include Enterobacteriaceae with resistance to carbapenem or those with carbapenemase production. Previous definition targeted carbapenemase-producing CRE and included resistance to carbapenems (excluding ertapenem) and third-generation cephalosporins^{Footnote r}. In 2013, the US CDC declared that the CRE threat level was urgent (immediate public health threat requiring aggressive intervention). In the United States, in the first half of 2012, at least one serious CRE case was found in 4% of the short-stay hospitals. On the other hand, 18% of the long-term acute-care hospitals had one case. Each year, almost 600 deaths result from carbapenem-resistant Klebsiella spp. and E. coli (the most common types of CRE in the United States. In 2012, 11% of Klebsiella spp. and 2% of E. coli were resistant to carbapenem in United States hospitals^{Footnote i}.

As per the EARS-NET 2014 report, carbapenem resistance in Europe was mediated by metallo-beta-lactamases (e.g., Verona integron-encoded metallo-β-lactamase and NDM enzymes) or serine-Carbapenemase (e.g., KPC enzymes)^{Footnote q}. It has been shown that CRE strains carrying plasmid-encoded carbapenemase enzymes like NDM and KPC have the ability to rapidly disseminate and therefore represent a public health concern^{Footnote s}.  Additionally, OXA-48-like enzymes, a family with carbapenemase-like activity, have also been found to reduce carbapenem susceptibility^{Footnote q}. The latter family has shown great ability to spread and cause hospital outbreaks. This can be explained by the high-transfer efficiency of the plasmid-containing OXA-48-like genes.

In the United States, from 2010 to 2015, 52 CRE isolates (collected from 43 patients) produced OXA-48-like carbapenemase, K. pneumoniae isolates which accounted for 81% of patients followed by E. coli isolates (16% of patients)

Footnote s

. In Europe, only three countries (Iceland, Montenegro and the former Yugoslav Republic of Macedonia) reported no CPE cases in 2013. Greece, Italy and Malta reported that CPE are regularly isolated from patients in most hospitals, corresponding to an endemic situation

Footnote t

.

Methicillin-resistant Staphylococcus aureus

Staphylococcus aureus (S. aureus) is a common gram-positive bacterium normally found on the skin of healthy individuals. The bacteria can also be found in the interior of the nose and the groin region of healthy individuals without causing disease (colonization). S. aureus can cause a variety of infections, most notably in the skin and soft tissue, bone and bloodstream (BSI). S. aureus that have acquired resistance to β-lactam antimicrobials (e.g., penicillins such as oxacillin, methicillin and dicloxacillin) are called methicillin-resistant Staphylococcus aureus (MRSA)^{Footnote u}.

S. aureus, including MRSA is one of the most common causes of healthcare-associated infections in Canada and is spread from person to person through direct contact and by contact with contaminated surfaces. Initially, MRSA was found mainly in healthcare settings but over the past decade, community-associated MRSA (CA-MRSA) has been increasing in Canada, especially in certain populations (e.g., Indigenous peoples, homeless people and intravenous drug users). MRSA can cause serious infections such as BSI in the hospital setting, which can lead to death. In the community it causes mostly skin and soft tissue infections.

Methods

PHAC has collected information on MRSA infections in hospitalized patients through CNISP since 1995. Hospitals participating in CNISP report screening specimens, clinical isolates (anatomical sites other than blood) and blood isolates. Screening specimens represent colonized cases and are isolated from nose, perineal, groin, axillary or other screening sites. Colonizations are reported from each site as an aggregate number with information on where MRSA was acquired. Individuals from whom MRSA was recovered from a clinical isolate or a positive blood culture are classified as either a clinical infection or bacteremia respectively. Case information include demographics and clinical information, previous hospitalization within the past 12 months, site of positive culture, where MRSA was presumed to have been acquired (either in the community or in the hospital), and outcome within the first 30 days following positive identification of MRSA culture. One MRSA clinical isolate for each clinically infected case collected between January 1^st and March 31^st of each calendar year are sent by participating hospitals to NML for molecular testing. One blood isolate from cases occurring at any time during the surveillance year are also sent to the NML for molecular testing.

MRSA in Canada

Data on MRSA infections in hospitalized patients presented here were reported to CNISP and were further classified into healthcare-associated (HA), community-associated (CA) and MRSA blood stream infections (BSI). In 2014, the rate of all MRSA infections in patients was 2.12 per 1,000 patient admissions and 2.89 per 10,000 patient days (Figure 12). Overall MRSA infection rates continued to rise from 1995 until 2008. Since 2008, there has been a decrease of approximately 25% in MRSA infection rates (2.92 per 1,000 patient admissions and 3.64 per 10,000 patient days in 2008).

Since 2008, the most dramatic reduction seen in MRSA infection rates was among those infections identified as healthcare-associated. HA-MRSA infection rates continued to rise from 1995 until about 2008; however, since 2008 there has been approximately a 30% decrease in HA-MRSA infections rates. In 2008, HA-MRSA infection rates were 2.02 per 1,000 patient admissions and 2.52 per 10,000 patient days in 2008 and in 2014 were 1.26 per 1,000 patient admissions and 1.72 per 10,000 patient days respectively (Figure 13).

CA-MRSA infection rates remained relatively low until 2005 when a dramatic increase (greater than 400%) was observed. In 2004, CA-MRSA rates were 0.21 per 1,000 patient admissions and 0.28 per 10,000 patient days and in 2005 they were 0.79 and 0.93 respectively. However, since 2005 CA-MRSA infection rates have remained relatively consistent (Figure 14).

National trends of MRSA-BSI infection from 1995 to 2014 saw rates slowly increasing until 2009. Since then MRSA-BSI rates have remained relatively stable (Figure 15).

In 2014, 444 (22%) of MRSA infections were from blood and 1,537 (78%) were from clinical sources other than blood. Skin and soft tissue infections (SSTI) represented the largest proportion (728, 43%) of MRSA infections, followed by surgical site (214, 37%), and respiratory (297, 15%) identified from clinical sources other than blood—37% when blood isolates were included.

The majority of cases were seen in those ≥ 25 years. Males with MRSA infections represented 59% of all cases. (Figure 16)

Among the patients with reported MRSA infections, 80% of all hospitalized patients with reported MRSA infections in 2014 survived; whereas, only 10% of the patients had died at 30 days following diagnosis. However, of those patients identified as having a MRSA-BSI, 23% had died at 30 days following diagnosis with MRSA-BSI compared to only 6% of those with a clinical (non-blood) infection. Death is reported as ‘all-cause mortality’ and not necessarily attributable to the MRSA infection. Therefore, it is not known whether the deaths observed were directly related to their MRSA infection.

Table 1 describes MRSA epidemic strain types identified from blood and clinical non-blood isolates submitted annually for cases with a clinical/blood MRSA infection. Three strain types, CMRSA-2, CMRSA-10 and CMRSA-7 together accounted for almost 90% of all strain types identified over the seven-year surveillance period. CMRSA-2 (corresponds to USA100/800) is most commonly attributed as a healthcare-associated genotype while CMRSA-7 (corresponds to USA400) and CMRSA-10 (corresponds to USA300) are attributed to community-associated genotypes. Between 2008 and 2014, CMRSA-2, the strain type most typically associated with hospital settings, remains the most predominant strain type identified nationally in both clinical and blood isolates followed by CMRSA-10 and CMRSA-7, the two strain types most commonly associated with community settings. In both clinical and blood isolates the proportion of CMRSA-10 has been steadily increasing since 2008.

Although intravenous vancomycin is one of the primary antimicrobials recommended for treatment of MRSA, both linezolid (IV and oral) and daptomycin (IV) are used within Canadian healthcare settings. Other antimicrobials that are treatment options for specific sites of MRSA infections include: telavancin (intravenous [IV]) for SSTIs, doxycycline for SSTIs, Trimethoprim-sulfamethoxazole (TMP-SMX) (IV) as well as clindamycin (IV and oral) for treating SSTIs and pneumonia^{Footnote v}.

In rare cases, S. aureus may become resistant to vancomycin, the antimicrobial most frequently used to treat serious MRSA infections. This leaves few treatment options available as vancomycin-resistant S. aureus (VRSA) identified to date were also resistant to oxacillin and other classes of antimicrobials. Although VRSA has been identified in the United States and the United Kingdom, there have been no identified cases in Canada to date^{Footnote w}. There have also been no documented resistance to tigecycline, linezolid or daptomycin in the CNISP isolates tested from 2008 to 2013. The proportion of tested MRSA isolates across Canada that were resistant to ciprofloxacin, erythromycin and clindamycin has remained relatively unchanged over this period (Table 2). There has been a slight decrease in the proportion of tested isolates that were resistant to tetracycline and TMP-SMX.

Antimicrobial susceptibility patterns can vary widely by geographical region. In Canada, resistance in MRSA to ciprofloxacin, erythromycin and clindamycin is slightly higher in the eastern regions, while resistance to tetracycline and TMP-SMX is lower in the east compared to the rest of Canada^{Footnote x}.

International perspective

In the United States, invasive MRSA infections are monitored through the EIP which is an active, population- and laboratory-based surveillance system across diverse United States geographic locations in addition to NHSN which collects hospital data^{Footnote y}. The United States rates are reported using population estimates unlike Canada’s rates that are reported in hospital patient admissions and patient days.

Comparing the Canadian MRSA rates to those in the United States is difficult due to use of different methodologies; however, the overall declining trend noticed in Canada has also been reported in the United States. The healthcare-associated^{Footnote 19} United States MRSA rates (both hospital and community onset) dropped by 32.5 % from 2007 to 2008 (27.1 cases per 100,000 population) until 2013 (18.3 cases per 100,000 population)^{Footnote z}. For MRSA bacteremia, the Canadian MRSA-BSI has remained relatively stable since 2009, whereas, MRSA-BSI were reduced by 73% from 2001 to 2009 in intensive care unit (ICU) patients with central lines in the United States^{Footnote aa}. Also, an overall 8% decrease in the United States hospital-onset MRSA bacteremia was reported between 2011 and 2013^{Footnote ab}.

In contrast to Canada and the United States, MRSA cases have been increasing elsewhere. In Denmark, for example, the highest reported rate for new MRSA cases^{Footnote 20} was in the year 2014 (52.7 per 100,000 population). The number of new MRSA cases in 2014 (n=2,965) represented a 42% increase compared to 2013 (n=2,092) and was as five times as in 2007 (n=663). In 2014, MRSA bacteremia in Denmark comprised 2.9 % out of all S. aureus bacteremia increasing from 1.7% in 2013. It is of note that Denmark reports on MRSA bacteremia only (blood isolates only), this has been voluntarily up until 2006^{Footnote ac}.

In Denmark, the highest proportions of drug resistance were seen in penicillin (77%), fusidic acid (15%), erythromycin (8%), clindamycin (8%) and norfloxacin (6%) in 2014. Additionally, between 2005 and 2014, resistance to erythromycin (5% to 8%), clindamycin (4% to 8%), fusidic acid (10% to 15%) and norfloxacin (3% to 6%) has increased^{Footnote ac}. On the other hand, in Canada, the top four drugs to which resistance was reported from MRSA blood isolates were cefoxitin (99.6%), erythromycin (87.7%), ciprofloxacin (85.4%) and clindamycin (39.6%) in 2013. Also, resistance to S. aureus bacteremia has increased for ciprofloxacin, erythromycin and fusidic acid from 83.8%, 84.2% and 8.1% in 2008 to 85.4%, 87.7% and 11.5% in 2013 respectively. Resistance to clindamycin has decreased from 65% to 39.6% within the same time period. Resistance to linezolid was not seen across the surveillance periods in either country.

Vancomycin-resistant Enterococci

Enterococci are part of the normal intestinal flora of both humans and animals, but may also cause a range of illnesses. Many infecting strains originate from the patient's intestinal flora. From here, they can spread and most commonly cause urinary tract infections (UTIs), BSIs, intra-abdominal infections and surgical wound infections in hospitalized patients. Enterococcus faecalis (E. faecalis) and Enterococcus faecium (E. faecium) are the most prevalent species cultured from humans, accounting for more than 90% of clinical isolates^{Footnote ad}.

The naturally high level of antimicrobial resistance found in Enterococci makes infections difficult to treat^{Footnote ac}. The acquisition of vancomycin resistance by Enterococci has affected treatment options and fueled debate regarding screening and infection control practices for this organism^{Footnote ae,Footnote af}. VRE, particularly E. faecium strains, are frequently resistant to all or a majority of antibiotics that are typically effective treatment for vancomycin-susceptible Enterococci. This leaves clinicians treating VRE infections with limited therapeutic options.

Methods

PHAC has been collecting information on hospitalized patients with VRE^{Footnote 21} through CNISP since 1999. Infection Control Professionals complete a standardized patient questionnaire through concurrent or retrospective chart review once a VRE is identified by the hospital laboratory. The questionnaire includes patient demographics and clinical information, previous hospitalization within the past 12 months, site of positive culture, where VRE was presumed to have been acquired (community or hospital), and whether the patient had concurrent infection with methicillin-resistant Staphylococcus aureus (MRSA). All specimens of VRE infected cases are sent by participating hospitals to NML for molecular testing. Polymerase chain reaction (PCR) was used to determine the presence of vancomycin resistant genes van A, B, C, D, E, G and L. Multilocus sequence typing and broth microdilution using Gram-Positive Sensititre panels were completed only for bloodstream infection specimens to determine genetic relatedness and antimicrobial susceptibility, respectively.

VRE in Canada

Between 1999 and 2007, the infection rates of VRE remained relatively stable at 0.10 cases per 10,000 patient days. Between 2008 and 2012, the rates began a steady increase, reaching a peak of 0.61 cases per 10,000 patient days in 2012. In 2014, the rate settled back to 0.45 cases per 10,000 patient days (Figure 17).

In 2014, 54 hospitals reported a total of 294 VRE infections to PHAC for an infection rate of 0.45 per 10,000 patient days. The molecular characterization of the infecting organism was known for 60% (n = 175) of the reported cases and of these 175 cases, 99% (n = 173) were E. faecium and 1% (n=2) were E. faecalis. Van A was the predominant gene identified in 98% (n=172) of the VRE infections.

Of the 294 reported cases, 85% were healthcare-associated within the reporting hospital. The remaining 15% were distributed between, healthcare-associated - any other healthcare facility/setting (outside of the reporting hospital), community-associated and unknown.

Figure 18 identifies the site of VRE infection for cases reported in 2014. It shows that BSIs were the most frequently-reported site, followed closely by infections identified in urine. Previously reported data indicate that VRE BSI rates are slowly increasing over time.

In 2014, VRE was recovered from multiple sites for 16 cases. Blood in combination with another site was the most common presentation for individuals infected at two or more sites. Of the cases of reported VRE, 52% percent (n=153) of the reported cases were over the age of 65 and 1% (n=3) was less than 19 years of age. Males accounted for 52% (n=152) of reported VRE infections.

Susceptibility results for 12 antimicrobial agents were available for 368 cases diagnosed with VRE BSI between 2009 and 2014 (Figure 19). The agents reported include: ampicillin, ciprofloxacin, erythromycin, gentamicin 500, levofloxacin, linezolid, nitrofurantoin, penicillin, quinupristin/dalfopristin, rifampin, high-level streptomycin and tetracycline.

Of the cases tested between 2009 and 2014, nearly all were resistant to ampicillin, ciprofloxacin, erythromycin, levofloxacin and penicillin. The proportion of cases exhibiting resistance to these antibiotics has remained stable over the surveillance period. Between 2009 and 2013, resistance to rifampin has been slowly increasing; however, the proportion fell from 95% in 2013 to 78% 2014. Since 2011, resistance to nitrofurantoin has also been steadily increasing from 8% of cases in 2011 to 20% in 2014.

Between 2009 and 2010, there was a noticeable decrease in the proportion of cases exhibiting resistance to streptomycin, however, since 2010, resistance has remained consistent. Resistance to tetracycline has fluctuated between 38% of cases to 54% of cases over the surveillance period. Gentamicin 500 resistance peaked in 2012 with 23% of tested cases exhibiting resistance, however by 2014, the proportion fell to 10%.

Only a few cases were found to be resistant to linezolid and quinupristin/dalfopristin between 2009 and 2014. Out of the 368 cases tested for resistance, one was determined to be resistant to linezolid and ten were resistant to quinupristin/dalfopristin.

For 2014, outcome at 30 days following the date of positive blood culture was unknown or the information was not provided for 3% (n= 2) of VRE BSI cases. Of the remaining 62 cases for which data were available, 36% (n=23) of patients were still alive and in hospital and 34% (n=22) had been discharged or transferred at 30 days follow-up. Twenty-seven percent (n = 17) of VRE BSIs reported to CNISP in 2014 died.

International perspective

Since the first description of VRE in a clinical isolate in Europe in 1988, VRE are increasing in prevalence worldwide. According to the National Healthcare Safety Network from 2009 to 2010, 35.5% of enterococcal hospital-associated infections were resistant to vancomycin, ranking as the second most common cause of nosocomial infections in the United States^{Footnote ag}. Each year, around 20,000 VRE cases are reported among inpatients and result in approximately 1,300 deaths^{Footnote i}.

In 2012, a majority of European countries reported vancomycin resistance in E. faecium below 5%, whereas only few countries reported estimates above 10%^{Footnote ah}. As per EARSS-NET, the mean VRE proportions increased from 6.2% in 2011 to 7.9% in 2014 in Europe. In 2014, resistance frequency ranged from 0% in Estonia, Finland, Iceland and Malta to 45.1% in Ireland. Increasing trends were seen in Bulgaria, Croatia, Denmark, Hungary, Ireland, Italy, Slovakia and United Kingdom from 2011 until 2014^{Footnote q}.

Most of the global VRE cases result from van A which is mostly carried by E. faecium

Footnote ai

.

Drug-resistant Streptococcus

The bacterium Streptococcus pneumoniae is the cause of invasive pneumococcal disease (IPD), a severe form of infection that can lead to pneumonia with bacteremia and meningitis. The burden of the disease is highest in young children and the elderly. S. pneumoniae can also cause otitis media (middle ear infection), sinusitis, peritonitis and rare cases of endocarditis^{Footnote aj}. Prevention of infection by some serotypes of S. pneumoniae can be achieved by immunization with pneumococcal vaccines^{Footnote ak}. Streptococcus pyogenes can cause invasive group A streptococcal diseases (iGAS) such as streptococcal toxic shock syndrome, necrotizing fasciitis (“flesh-eating” disease) and bacterial sepsis, as well as non-invasive diseases like pharyngitis (“strep throat”), scarlet fever, rheumatic fever and skin infections such as impetigo^{Footnote aj}.

Methods

Provincial public health laboratories submit invasive Streptococcus isolates to the NML for serotyping, including information on patient age, gender, clinical source and date of collection; whereas the Laboratoire de santé publique du Québec, Alberta Provincial Laboratory for Public Health and the Toronto Invasive Bacterial Diseases Network submit serotyping data on invasive S. pneumoniae isolates that have been identified in their laboratories. The data were aggregated by age into <2 year, 2-14 year, 15-49 year, 50-64 year and ≥65 year age groups; and regionally into Western (British Columbia, Alberta, Saskatchewan, Manitoba, Yukon Territories, Northwest Territories and Nunavut); Central (Ontario and Quebec) and Eastern (New Brunswick, Nova Scotia, Prince Edward Island, Newfoundland and Labrador) regions of Canada. Submission of isolates is voluntary and not standardized across the country.

In 2011, the NML began a collaboration with the University of Manitoba – Health Sciences Centre - Canadian Antimicrobial Resistance Alliance (CARA) to provide antimicrobial susceptibility testing (AST) for S. pneumoniae isolates submitted called SAVE (S. pneumoniae Serotyping and Antimicrobial Susceptibility: Assessment for Vaccine Efficacy in Canada After the Introduction of PCV13).  All sterile-site isolates from any age group causing invasive pneumococcal disease submitted by 8 participating jurisdictions (Saskatchewan, Manitoba, Ontario, Quebec, Nova Scotia, Prince Edward Island, New Brunswick, Newfoundland and Labrador) are included in the study.

Drug-resistant S. pneumoniae in Canada

In Canada, since the implementation of routine immunization programs with 13-valent pneumococcal conjugate vaccine (PCV13) beginning in 2010, the national annual incidence rate of IPD has decreased from 9.6 cases per 100,000 population in 2011 to 8.9 cases per 100,000 population in 2014. During this time, the average annual incidence rate was highest in people aged 60 years and older (21.8 cases per 100,000 population), followed by infants less than one year (18.4 cases per 100,000 population) and children aged one to four years (12.8 cases per 100,000 population).

Resistance of S. pneumoniae to penicillin has decreased from 12% in 2011 to 9% in 2014 (Figure 20) and resistance to clindamycin declined from 7% to 4% over the same period. Resistance to doxycycline (8%) has remained relatively stable since 2010. Resistance to clarithromycin, which can be used in community-acquired pneumonia, decreased from 25% in 2013 to 22% in 2014 and TMP-SMX resistance has remained stable at 6%. To date, there has been no resistance reported to linezolid, tigecycline or vancomycin. Carbapenem resistance among S. pneumoniae is less common being detected at levels less than 2% in 2014.

Beginning in 2010, PCV13 was introduced in Canada, which included protection against the multidrug-resistant (MDR) serotype, 19A. The vaccine-induced decline of PCV13 serotypes (including the MDR serotype 19A) has driven a concurrent decline in overall AMR in pneumococci (Figure 21)^{Footnote al}. PCV13 serotypes have declined from 50% of the isolates in 2011 to 31% in 2014. Resistance to three or more classes of antimicrobials has also declined from 8% to 5% over the same time period. The impact of the pneumococcal vaccine on resistant strains illustrates the importance of vaccines as part of a strategy to mitigate the impact of AMR.

Drug-resistant S. pyogenes in Canada

From 2009 to 2013, the national incidence rate of iGAS in Canada has increased significantly (p<0.0001) from 4.0 to 4.7 cases per 100,000 population with an average annual incidence rate of 4.5 cases per 100,000 population (range: 4.0-4.9). The average annual incidence rate per 100,000 population was highest in infants less than one year of age (9.4 cases, range: 8.7-9.8), followed by the over 60 age group (7.2 cases, range: 6.6-7.6).

The most predominant types of iGAS in 2014 were emm types 1 and 89 accounting for 28% and 10% of the isolates tested, respectively. The majority of iGAS was isolated from blood samples (69%), followed by synovial fluid (8%). All iGAS were susceptible to penicillin and vancomycin. Resistance to clindamycin, a second-line drug for treatment, has remained relatively unchanged from 2010 to 2014 with about 2% of isolates resistant to this antimicrobial, while resistance to erythromycin and non-susceptibility to chloramphenicol has decreased from 14% to 7% and 9% to 0.1%, respectively, over the same time period (Figure 22).

International perspective

Similar to Canada, the overall IPD rate has declined in the United States from 2011 to 2014 (11.8 cases per 100,000 population to 8.7 cases per 100,000 population, respectively). In the United States, IPD rates in people ≥65, compared to the relatively stable rates in Canada, have increased from 35 cases per 100,000 population to 40.9 cases per 100,000 population between 2011 and 2014. As seen in Canada, for children less than five years, the United States rates have decreased from 12 cases per 100,000 population in 2011 to 8.4 cases per 100,000 population in 2014.

In the United States, between 2011 and 2014, resistance to erythromycin was detected in 28.4% of the overall IPD isolates increasing from 26.2% in 2011 to 30.2% in 2014. Tetracycline resistance was lower at 11.5% decreasing from 13.2% in 2011 to 11.7% in 2014. TMP-SMX resistance was on average 10.7% decreasing from 14.3% in 2011 to 9.7% in 2014^{Footnote am,Footnote an}.

The top three drugs to which resistance was found in Canada were clarithromycin (23.2%) followed by penicillin (10.4%) and then doxycycline (9.3%). In 2014, resistance to penicillin in Canada (8.6%) was higher compared to the United States (2%) and the United Kingdom (0.5%)^{Footnote ao}.

Invasive GAS rates were stable in the United States from 2009 until 2013 (3.6 to 3.7 cases per 100,000 population) in contrast to the significant increase seen in Canada^{Footnote ap,Footnote aq}. Yet, in 2014, United States rates increased by 13% to 4.4 cases per 100,000 population^{Footnote ar}.

The 2013 US CDC antibiotic resistance threats report indicated that from 2010 to 2011; 10% of the GAS isolates were erythromycin-resistant^{Footnote i}. From 2010 to 2014 a decline in iGAS resistance to erythromycin (14.4% to 6.9%) was seen in Canada. Clindamycin resistance was higher in the United States (3.4%) compared to Canada (2%) from 2010 to 2011.

Neisseria gonorrhoeae

Neisseria gonorrhoeae (N. gonorrhoeae) causes gonorrhea, a highly infectious sexually transmitted infection. It commonly results in genital infection that may be symptomatic or asymptomatic. Other sites of infection are also possible. If untreated or inappropriately treated, it may cause genital/reproductive tract inflammation and damage as well as infertility. The treatment and control of gonorrhea is complicated because N. gonorrhoeae develops resistance to the antimicrobials used to treat it, including penicillins, tetracyclines, macrolides and quinolones. Recently, isolates with resistance to azithromycin and decreased susceptibility to cephalosporins are emerging and threatening the last available treatment options. The Canadian Guidelines on Sexually Transmitted Infections updated recommendations for the use of combination gonorrhea therapy with 250 milligrams ceftriaxone intramuscularly and azithromycin 1 grams orally as the first-line treatment for uncomplicated anogenital and pharyngeal infections in adults^{Footnote as}.

Methods

Provincial public health laboratories submit N. gonorrhoeae isolates to the NML as part of the passive National Neisseria gonorrhoeae Surveillance Program, including information on age and gender of the patient as well as the anatomical site of infection. Laboratories submit isolates when resistance to one antimicrobial has been identified or if the provincial laboratories do not perform any antimicrobial susceptibility testing. Submission of isolates is voluntary and is not standardized across the country. Total number of isolates cultured in all provinces was used as the denominator to calculate resistance proportion. Antimicrobial susceptibility testing was conducted using agar dilution for the following antimicrobials: penicillin, tetracycline, erythromycin, spectinomycin, ciprofloxacin, ceftriaxone, cefixime, azithromycin, ertapenem and gentamicin.

Gonorrhea in Canada

Gonorrhea is the second most commonly reported bacterial sexually transmitted infection in Canada. Between 2004 and 2013, the rate of reported cases of gonorrhea increased by 43.1%, from 27.4 to 39.3 per 100,000 population, particularly in females. In 2013, as in previous years, the rate of reported cases of gonorrhea was higher in males than females (47.4 vs. 31.1 per 100,000 population). Females between the ages of 15 and 24 years and males between the ages of 20 and 29 accounted for the highest rates of gonorrhea in 2013^{Footnote at}.

Of the 3,809 N. gonorrhoeae isolates cultured in public health laboratories across Canada in 2014, a total of 1,955 (52.4%) were found to be resistant to at least one antibiotic tested^{Footnote au}.

Isolates with decreased susceptibility to ceftriaxone (minimum inhibitory concentration [MIC] greater or less than 0.125 milligrams per litre) have declined from 7.3% (n=218/2,970) in 2010 to 2.7% (n=101/3,809) in 2014. Those with decreased susceptibility to cefixime (MIC greater or less than 0.25 milligrams per litre) have declined from 4.2% (n=140/3,360) in 2011 to 1.1% (n=42/3,809) in 2014. The proportion of azithromycin-resistant (MIC greater or less than 2 milligrams per litre) N. gonorrhoeae isolates increased from 0.4% (n=11/3,106) in 2009 to 3.3% (n=127/3,809) in 2014, including 38 isolates from an outbreak (Figure 23). In 2014, 34.0% (n=1,296/3,809) of the isolates were resistant to ciprofloxacin; 32.0% (n=1,219/3,809) were resistant to erythromycin; 18.2% (n=693/3,809) were resistant to penicillin and 47.3% (n=1,809/3,809) were resistant to tetracycline (Figure 24)^{Footnote au}.

Since 2012, isolates with resistance to azithromycin and decreased susceptibility to cephalosporins (cefixime and ceftriaxone) have been observed in N. gonorrhoeae isolates in Canada, with a total of 0.2% (n=7/3,036) in 2012, 0.3% (n=8/3,195) in 2013 and 0.03% (n=1/3,809) in 2014. Despite the small numbers, this is of concern as it represents a threat to the success of currently recommended dual therapy treatment options^{Footnote au}.

International perspective

The decrease in the proportion of isolates with elevated MICs to the cephalosporins observed in Canada was also reported in the United States and the United Kingdom. In the United States isolates with decreased cefixime susceptibility declined from 1.4% in 2011 to 0.4% in 2013 and decreased ceftriaxone susceptibility declined from 0.4% in 2011 to 0.05% in 2013^{Footnote aw}. The United Kingdom reported that the proportion of isolates with decreased cefixime susceptibility declined from 6.3% in 2010 to 1.3% in 2013 and decreased ceftriaxone susceptibility declined from 0.3% in 2009 to 0.1% in 2013^{Footnote ax}. In Canada, N. gonorrhoeae azithromycin resistance levels were higher than in the United States (MIC greater or less than 2 milligrams per litre), which ranged from 0.2% to 0.6% between 2009 and 2013 and in the United Kingdom (MIC greater or less than 1 milligrams per litre), which reported 1.6% in 2013^{Footnote ay,Footnote az}. Australia reported 2.1% azithromycin resistance among their isolates in 2013^{Footnote ax}.

Drug-resistant Salmonella Typhi

Enteric fever is caused by Salmonella serotypes Typhi and Paratyphi. It is an enteric febrile illness characterized by fever, rash, diarrhea (more common in children) or constipation (more common in adults). Children usually present with milder symptoms compared to adults. Serious systemic manifestations can also occur (e.g., myocarditis). Three percent of patients might present with bleeding due to intestinal perforation^{Footnote ay}.

Humans are the only reservoir for Typhoidal Salmonella. Infection usually occurs from consumption of food or water that has been contaminated by an ill person or a chronic asymptomatic carrier^{Footnote az}. In Canada, enteric fever is usually acquired during travel and the destination of travel is the strongest predictor of S. Typhi risk with travel to South Asia^{Footnote 23} posing the highest risk.

The first line for empiric therapy is a fluoroquinolone with ciprofloxacin being the most commonly used^{Footnote ba}. However, when deciding on the optimal empiric therapy, antimicrobial resistance patterns in the travel destination countries should be considered^{Footnote ay}. When fluoroquinolone resistance is suspected, injectable third-generation cephalosporins are the empiric treatment of choice. Azithromycin is being increasingly used to treat enteric fever because of the emergence of multidrug-resistant strains^{Footnote ba}.

Methods

Provincial public health laboratories submit all Salmonella Typhi, Paratyphi A and Paratyphi B to the NML for antimicrobial susceptibility testing, including age and gender of the patient as well as the anatomical site of the infection. The Yukon, Northwest Territories and Nunavut forward their isolates to one of the provincial laboratories. Laboratories submit isolates as part of the Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS).  Antimicrobial drug susceptibility testing was performed using broth microdilution and breakpoints established by the Clinical Laboratory Standards Institute whenever available. The antimicrobials in the susceptibility panel include the following antimicrobials: amoxicillin-clavulanic acid, ceftiofur, ceftriaxone, ciprofloxacin, ampicillin, azithromycin, cefoxitin, gentamicin, kanamycin, nalidixic acid, streptomycin, trimethoprim-sulfamethoxazole, chloramphenicol, sulfisoxazole and tetracycline.

Drug-resistant Salmonella Typhi in Canada

In Canada, the rate of typhoidal Salmonella infection has ranged from a low of 0.5 cases per 100,000 population in 2003 to a high of 0.7 cases per 100,000 population in 2008. Between 2009 and 2011, rates of S. Typhi infections have been very low in Canada ranging between 0.39 and 0.53 cases per 100,000 population respectively^{Footnote bb}. A total of 184 typhoidal isolates were tested for antimicrobial susceptibility in 2014 by the Canadian Integrated Programs for Antimicrobial Resistance Surveillance (CIPARS); S. Typhi (n=148), S. Paratyphi A (n=29) and S. Paratyphi B ^{Footnote 24} (n=7). The majority of these typhoidal isolates tested in 2014 were from residents of Ontario, British Columbia and Alberta.

Antimicrobial resistance for Salmonella (typhoidal and non-typhoidal serotypes) has been monitored by CIPARS since 2002. Similar to previous years, the majority of typhoidal isolates (82%) were resistant to nalidixic acid (Figure 25). Ciprofloxacin resistance has increased significantly in Canada from zero percent in 2003 to 17% in 2013 and 14% in 2014. No resistance to ceftriaxone or azithromycin was reported in 2014. A total of 18% (33/184) of isolates were susceptible to all antimicrobials tested whereas 16% were multiclass-resistant (resistant to ≥ 3 classes of antimicrobials).

International perspective

In 2014, the World Health Organization (WHO) estimated that almost 21 million cases of S. Typhi infections with related 222,000 deaths occur worldwide annually^{Footnote bc}. Almost 90% of these deaths occur in Asia. The annual incidence rate of blood culture-confirmed cases in Asian urban slums is estimated to be 180–494 per 100,000 for children between five and 15 years old and similar or higher rates occur in children younger than five years old in these areas^{Footnote bd}. Resistance to nalidixic acid and decreased susceptibility to ciprofloxacin is becoming endemic in the Indian subcontinent and in Southeast Asia^{Footnote be}.  

In the United States, around 300 culture-confirmed cases of S. Typhi and 100 cases of S. Paratyphi serotypes are reported annually. As seen in Canada, the vast majority of cases (greater than 80% of S. Typhi and greater than 90% of S. Paratyphi A) are among travelers to South Asia. On the other hand, the risk of S. Typhi infection has declined in some travel destination countries. As a result, US CDC removed pre-travel typhoid vaccination recommendations for travel to Eastern Europe and the Middle East

Footnote ba

. Resistance to S. Typhi has been declared a serious threat by the US CDC. Resistance to ciprofloxacin has increased significantly in the United States from 20% in 1999 to more than 70% in 2011. From 2009 to 2011, approximately 67% of S. Typhi infections were resistant or partially resistant to ciprofloxacin. On the other hand, US CDC reports that until 2013 no resistance to azithromycin or ceftriaxone was seen in the United States for S. Typhi

Footnote bf

.

Drug-resistant Non-typhoidal Salmonella

Methods

Provincial public health laboratories submit all Salmonella, Paratyphi A and Paratyphi B to the NML for antimicrobial susceptibility testing, including age and gender of the patient as well as the anatomical site of the infection. The Yukon, Northwest Territories and Nunavut forward their isolates to one of the provincial laboratories. Laboratories submit isolates as part of the Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS). 

Antimicrobial drug susceptibility testing was performed using broth microdilution and breakpoints established by the Clinical Laboratory Standards Institute whenever available. The antimicrobials in the susceptibility panel include the following antimicrobials: amoxicillin-clavulanic acid, ceftiofur, ceftriaxone, ciprofloxacin, ampicillin, azithromycin, cefoxitin, gentamicin, kanamycin, nalidixic acid, streptomycin, trimethoprim-sulfamethoxazole, chloramphenicol, sulfisoxazole and tetracycline.

Drug-resistant non-Typhoidal Salmonella in Canada

In 2014, a total of 2,496 non-typhoidal Salmonella human isolates were submitted to PHAC for antimicrobial susceptibility testing, the majority of which were submitted by Ontario (35%) followed by Quebec (17%), British Columbia (11%) and Alberta (11%). Salmonella Enteritidis was the most common serotype associated with human disease submitted for susceptibility testing, followed by S. Heidelberg and S. Typhimurium.

The majority of the Salmonella isolates were recovered from stool samples (83%), followed by blood (5%) and urine (4%) (Table 3).

In 2014, 75% of all human non-typhoidal Salmonella isolates were susceptible to all antimicrobials tested, with 7% of isolates exhibiting resistance to three or more antimicrobial classes (multiclass-resistant) (Table 4). Ampicillin was the antimicrobial for which a larger proportion of isolates were resistant (8%), followed by tetracycline (7%), streptomycin (6%) and gentamicin (6%). No resistance to azithromycin was reported in 2014 and only 1% of isolates were resistant to ciprofloxacin, both antimicrobials used in treatment of severe human infections. Over time, resistance to most antimicrobials has been decreasing since 2004 (Figure 26), with the exception of resistance to nalidixic acid which has nearly doubled between 2013 (5%) and 2014 (9%). Most of the nalidixic-resistant isolates from humans are Salmonella Enteritidis and the increase in resistance to this antimicrobial is likely associated with the large number of Enteritidis cases observed in 2014.

International perspective

The US CDC has reported that they have observed resistance to ceftriaxone in about 3% of non-typhoidal Salmonella tested and some level of resistance to ciprofloxacin in about 3%. Multidrug resistance (greater than five classes) has been observed in approximately 5% of non-typhoidal Salmonella tested by the US CDC^{Footnote i}.

Drug-resistant Tuberculosis

Tuberculosis (TB) is an infectious disease caused by Mycobacterium tuberculosis that primarily affects the lungs but can also affect any part of the body. It is transmitted by the inhalation of airborne infectious droplet nuclei produced by an individual with infectious pulmonary and/or laryngeal TB when coughing, sneezing, talking or spitting. A susceptible individual usually requires prolonged exposure before becoming infected. Adolescents and adults are most likely to transmit TB infection, although young children can sometimes be infectious^{Footnote bj}.

Individuals diagnosed with active TB disease are said to have drug-resistant TB if the strain ofTBcausing their disease is resistant to one or more of the four first-line drugs. The following resistance patterns are of concern: Mono-resistance—defined as resistance to one first-line anti-TB drug only (isoniazid, rifampin, ethambutol or pyrazinamide); Polyresistance (other patterns)—defined as resistance to more than one first-line anti-TB drug, not including the isoniazid and rifampin combination; Multidrug-resistant tuberculosis (MDR-TB)—defined as resistance to isoniazid and rifampin with or without resistance to other anti-tuberculosis drugs; and  Extensively drug-resistant TB (XDR-TB)—defined as resistance to isoniazid (INH), rifampin (RMP), any fluoroquinolone and at least one of the three injectable second-line drugs (amikacin, capreomycin or kanamycin).

TB is a major public health problem that affects millions of people around the globe, predominantly in low- and middle-income countries^{Footnote bk}. TB, however, remains a persistent health threat in high-income countries, particularly among the poorest, most vulnerable segments of the population^{Footnote bl}. Globally, the improper prescription of anti-TB drugs, their proper prescription but unavailability, inadequate supervision or, uncommonly, the malabsorption of these drugs has increased the prevalence of drug-resistant TB^{Footnote bj}.

Methods

PHAC collects information on tuberculosis in Canada through the Canadian Tuberculosis Laboratory Surveillance System (CTBLSS) and the Canadian Tuberculosis Reporting System (CTBRS).  CTBLSS is an isolate-based surveillance system developed to collect timely data on tuberculosis (TB) drug resistance across Canada.  Participating laboratories include members of the Canadian Tuberculosis Laboratory Technical Network representing all provinces and territories.  Participating laboratories submit drug susceptibility test results for all first-line TB drugs (isoniazid, ethambutol, rifampin, and pyrazinamide) and the results for at least one of the fluoroquinolones (levofloxacin, Moxifloxacin, or ofloxacin), and the injectable agents (amikacin, capreomycin and kanamycin).  CTBRS is a case-based surveillance system that maintains information on people diagnosed with active TB disease. Provincial and territorial public health authorities voluntarily submit data on all new and re-treatment cases of active TB disease that meet the Canadian case definition. Individuals diagnosed with active TB disease are said to have drug-resistant TB if the strain ofTBcausing their disease is resistant to one or more of the four first-line drugs.

Drug-resistant Tuberculosis in Canada

Canada has one of the lowest TB disease rates in the world^{Footnote bk}. Between 2004 and 2014, both the number of reported TB cases and the overall Canadian incidence rates have remained relatively stable (Figure 27). Overall, between 2004 and 2014, there were 17,902 cases of active TB disease reported to the CTBRS for an average of 1,627 cases reported annually. Over the same period, the incidence rate remained relatively stable between 4.4 cases to 5.0 cases per 100,000 population.

Between 2004 and 2014, 80% (14,338) of all reported TB cases in Canada were culture-positive. Of these 98% (14,047) were tested for resistance.  Nine percent (9%, 1,268 cases) of the culture-positive cases were found to be resistant to at least one of the first-line anti-tuberculosis drugs. Isoniazid resistance was the most common pattern of first-line drug resistance reported (Figure 27).

Between 2004 and 2014, of the culture-positive cases tested for resistance (n=14,047), 8% (n = 1,070) were monoresistant, 0.5% (n = 60) were identified as polyresistant, 1% (n=150) were MDR-TB and less than 0.1% (n=5) were XDR-TB (Figure 28). The percentage of cases with MDR-TB has remained between 1% and 2% for the period 2004 and 2014 (Figure 28).

Geographic distribution^{Footnote 25}

Data from the CTBLSS show that every province and territory in Canada reported at least one case of drug-resistant TB between 2004 and 2014^{Footnote bm}. As well, Alberta, British Columbia, Manitoba, Ontario and Quebec have all reported cases of MDR-TB. Of the five XDR-TB cases reported between 2004 and 2014, three were reported from Ontario and one each from Manitoba and Quebec.

Between 2004 and 2014, of the reported cases with drug resistant TB (i.e., resistance to at least one of the first-line antimicrobials), 46% (n= 585) were females and 54% (n = 683) were males. Ten percent of females and 9% of males tested had TB that was resistant to at least one of the first-line TB drugs. For both males and females, 1% of the cases tested had TB that was resistant to both INH and RMP. Among the XDR-TB cases, four of the five cases were female. For all cases, the majority, 42% (n = 541) were between the ages of 25 and 44.

Two percent (n = 69) of the Canadian-born Indigenous cases, 9% (n = 141) of the Canadian-born non-Indigenous cases and 11% (n = 1,047) of the foreign-born cases were resistant to at least one of the first-line antimicrobials (Figure 29). The foreign-born cases accounted for 83% (n=1,047) of all resistant cases, 97% (n=145) of the MDR-TB cases and four of the five XDR-TB cases.

Among the foreign-born, drug-resistant TB cases presented earlier after arrival in Canada than drug-susceptible TB cases (Figure 30). Of the susceptible cases for which year of arrival was recorded (n = 7,517), 42% arrived in Canada within five years of the TB diagnosis compared to 60% of MDR-TB cases (n = 144).

Between 2004 and 2014, drug-resistant TB was reported slightly more often in cases with a past history of TB (re-treatment cases) as compared with those with no prior history of disease (new cases). Of 12,901 new active TB cases reported, 0.9% (n = 114) were MDR-TB whereas for the 907 re-treatment case 3.8% (n = 35) were MDR-TB. This difference in proportions is significant (p < 0.001).

As routine testing of TB cases for resistance to fluoroquinolones (levofloxacin, moxifloxacin and ofloxacin) or injectable agents (amikacin, kanamycin and capreomycin) is not done in Canada, the epidemiology of second-line drug resistance has not been clearly described. Testing is typically considered where resistance to INH and RMP has been established. Of the 155 cases resistant to both INH and RMP, 142 were tested for resistance to at least one fluoroquinolone and one of the injectable agents. Of these 142, 10% (n = 15) were also resistant to a fluoroquinolone and 13% to at least one of the three injectable agents.

Final treatment outcome was assessed for cases diagnosed between 2006 and 2013. For these cases (n = 10,525) treatment outcome was reported for 92% (n = 9,730) of cases. Eighty-four percent of cases were sensitive to all four first-line medications, 83% of the monoresistant cases and 79% of the MDR-TB cases were reported as cured or had completed treatment. At the time the outcome data were reported to PHAC, 3% of the fully-sensitive cases were still on treatment whereas 11% of the MDR-TB cases were still on treatment. Death was recorded as the final outcome for 10% of the fully-sensitive cases and 4% of the MDR-TB cases.

The duration of treatment varied depending on the reported pattern of drug resistance. For cases that were fully sensitive (n = 7,332) the mean duration of treatment was 8.7 months (95% CL: 8.6 – 8.8 months). For those cases with any resistance, excluding the MDR-TB cases, (n = 588), the duration of treatment was 10.8 months. Finally, of those cases resistant to INH and RMP for ± 0.3 months which outcome data were available (n = 72) the mean duration of treatment was 23 months ± 1 month (95% CL: 21 – 24 months).

International perspective

The WHO estimates that, globally, in 2014, 9.6 million people fell ill with TB and 1.5 million of those with TB died. An estimated 480,000 people developed MDR-TB and an estimated 9.7% of people with MDR-TB had XDR-TB. An estimated 3.3% of new and 20% of previously treated cases had MDR-TB. These estimates remained unchanged from those reported in 2012. Globally, in 2014, of the individuals diagnosed with MDR-TB or rifampicin-resistant tuberculosis, 75% lived in the European Region, India, South Africa or China^{Footnote bk}.

In 2014, the reported TB incidence rate in the United States was 3.4 per 100,000 population, which was lower than the Canadian rate for the same year^{Footnote bn}. Between 2004 and 2014, in the United States, 1% of all reported new cases and 5% of re-treatment cases were diagnosed with MDR-TB. These percentages are comparable to the proportions reported in Canada. Between 2009 and 2014, there were 15 XDR-TB cases reported in the United States representing less than 0.1% of all reported culture-positive cases in that period. This is also similar to what was observed in Canada, where less than 0.1% (n=5) of cases reported between 2004 and 2014 were XDR-TB^{Footnote bo}.

The burden of TB in Canada among the foreign-born reflects the global TB patterns. Between 2004 and 2014, there were 17,902 reported cases of active TB disease in Canada, of which 67% (n=11,942) were foreign-born. Of all the foreign-born cases, 46% were from three countries, China, India and the Philippines—countries identified by the World Health Organization as High Burden TB countries. The remaining cases came from over 180 different countries. Of those reported cases with MDR-TB, 96% (n=149) were foreign-born and 56% (n=84) case were born in the Philippines (n=26) China (n=22), India (n=20) and Vietnam (n=16)^{Footnote bp}.

Resistance in foodborne bacteria

PHAC monitors antimicrobial resistance in selected bacterial organisms in food sources across Canada.  These include E. coli, Salmonella and Campylobacter species which are prevalent in food-animal sources and impact on human health. The contamination of animals and animal products with antimicrobial resistant bacteria has been identified as a source for human infection with resistant organisms and these organisms are a frequent cause of food-borne outbreaks.

Most individuals infected with food-borne E. coli, Salmonella and Campylobacter species will develop diarrhea, fever and abdominal cramps.  In most cases, the illness is self-limited and treatment is not required.  Some vulnerable individuals, such as the elderly, very young children, and individuals with underlying medical conditions; may need to be hospitalised if the diarrhea is severe.  Pregnant women are also at increased risk of complications related to these organisms.  Some strains of E. coli, such as E. coli O157:H7 or enterohemorrhagic E. coli (EHEC) can cause a life-threatening condition known as hemolytic uremic syndrome (HUS). Individuals with HUS can develop permanent kidney damage and potential death.

Methods

The Canadian Integrated Program for Antimicrobial Resistance Surveillance (CIPARS) monitors antimicrobial resistance in three zoonotic bacteria: generic Escherichia coli, Campylobacter and Salmonella. Samples are collected at three points along the food chain: 1) from healthy animals on farm, 2) from healthy animals at slaughter and 3) meat at retail food stores. CIPARS focuses sampling on the major meat commodities consumed in Canada: chicken, pork and beef. Table 5 indicates which zoonotic bacteria were isolated from which animal species along the food chain in 2014. Additionally, clinical Salmonella isolates from humans and animals are also tested for antimicrobial resistance.

Generic Escherichia coli

Chicken

Generic E. coli are found everywhere as they inhabit the gastrointestinal tract of animals and humans and therefore act as good indicators of antimicrobial resistance selection pressure. In 2014, a total of 1,294 samples from chickens were submitted for bacterial and antimicrobial susceptibility testing. The proportion of samples positive for generic E. coli has remained consistent, with 96% of all samples collected in 2014 found to be positive (Table 6).

Approximately half of the generic E. coli isolates obtained from chicken samples along the food chain were found to be resistant to tetracycline, followed by ampicillin (43%) and streptomycin (42%) (Figure 31). Resistance to ceftriaxone decreased between 2004 (31%) and 2014 (21%), while resistance to gentamicin and trimethoprim-sulfamethoxazole increased during this time period from 9% to 18% and 9% to 16%, respectively.

In 2014, all provinces/regions participating in the CIPARS broiler chicken farm program reported a decrease in the use of ceftiofur, a third-generation cephalosporin, compared to 2013. This corresponded with decreased levels of third-generation cephalosporin resistance among generic E. coli from broiler chickens. In 2014, the Canadian chicken industry implemented a policy eliminating the preventive use of antimicrobials considered of very high importance to human medicine, including third-generation cephalosporins, and this intervention is likely responsible for most of the decrease in third-generation cephalosporin use reported in broiler flocks.

Swine

Swine samples were collected at different stages along the food chain: sentinel grower-finisher swine farms, federally-inspected slaughter houses and at retail stores across the country. Over time the recovery rates for generic E. coli among swine samples collected along the food chain has ranged from 50% (2005) to 62% (2004), with 55% of all swine samples collected in 2014 found to be contaminated with this organism (Table 7).

Sixty-nine percent of all E. coli isolates from pigs were resistant to tetracycline in 2014. Two percent of isolates were resistant to ceftriaxone and gentamicin and less than 1% were resistant to nalidixic acid (Figure 32). Since surveillance of grower-finisher pigs on farm began in 2006, the level of resistance in E. coli from pigs has remained relatively stable.

Cattle

Samples cultured for E. coli were collected at the slaughter house and retail level. Currently there is no surveillance of cattle at the farm level. This should be taken into consideration when reviewing to recovery rates below (Table 8).

Similar to that observed among chicken and swine samples, the largest proportion of generic E. coli isolates were resistant to tetracycline, with 20% of all isolates resistant to this antimicrobial in 2014 (Figure 33). Very little (≤ 0.3%) resistance was observed to ceftriaxone and the antimicrobials gentamicin and nalidixic acid.

Campylobacter species

Chicken

Testing for Campylobacter spp. contamination is carried out on chicken samples collected from farms (broiler chickens), slaughter house and retail outlets across Canada. Total recovery rates are presented below (Table 9). In 2014, 26% (294/1149) of retail, 27% (187/683) of abattoir and 16% (93/564) of farm chicken samples were positive for Campylobacter.

In 2014, 44% of Campylobacter spp. isolates were resistant to tetracycline (Figure 34). Resistance to ciprofloxacin reported in 2014 (11%) was higher than resistance observed in 2004 (4%) and was consistent across all points along the food chain—retail (11%; 30/277), slaughter house (11%; 20/188) and farm (10%; 9/93).

Danofloxacin and enrofloxacin are fluoroquinolones approved for use in livestock in Canada, specifically for treating respiratory disease in cattle and pigs (there are no approved uses in chicken). When data about ciprofloxacin resistance in Campylobacter spp. recovered from chicken samples were examined at the regional level, it was noticed that trends varied considerably between provinces/regions and decreases in reported fluoroquinolone use in some provinces appeared to coincide with observed decreases in ciprofloxacin resistance.

Swine

Samples from pigs for Campylobacter spp. detection are only collected at the slaughter house level and testing for Campylobacter only started in 2012. Retail samples are not tested for Campylobacter due to the low levels detected during the beginning of surveillance. Campylobacter is commonly isolated from pigs at slaughter (Table 10); most of the isolates from pigs are Campylobacter coli.

In 2014, 78% of isolates were resistant to tetracycline, followed by azithromycin (53%) and telithromycin (43%). Eleven percent of isolates were resistant to ciprofloxacin (Figure 35).

Cattle

Campylobacter spp. have been isolated from cattle samples collected at the slaughter house level since 2008. Similar to Campylobacter spp. in swine, this organism is not tested for among beef samples collected at the retail meat level due to low detection levels identified at the beginning of the Surveillance Program. Recovery rates for Campylobacter spp. among cattle samples has ranged from 53% in 2010 to 92% in 2012, with 87% of all samples testing positive for Campylobacter spp. in 2014 (Table 11).

Antimicrobial resistance in Campylobacter spp. isolates recovered from cattle samples is mainly to tetracycline, with 54% of isolates resistant to this antimicrobial in 2014 (Figure 36). Resistance to azithromycin and telithromycin was reported for the first time in 2014, although less than 1% of isolates were resistant to these antimicrobials. Resistance to ciprofloxacin has increased between 2008 and 2014 from 2% to 7% during this time period.

Salmonella spp.

Chicken

Testing for Salmonella spp. is carried out on chicken samples collected from farms, slaughter and retailers. In 2014, 34% of all chicken samples collected in Canada were found to be contaminated with Salmonella spp. (Table 12). Among the Salmonella recovered from chicken, S. Enteritidis and S. Kentucky were the main serotype identified, with S. Heidelberg being the third most common serotype.

In 2014, 56% of all Salmonella isolates recovered from chicken along the food chain were fully susceptible to all antimicrobials tested, with 6% of isolates classified as multiclass-resistant (resistant to three or more antimicrobial classes). Thirty percent of chicken Salmonella isolates were resistant to tetracycline, followed by streptomycin (26%) and ampicillin, amoxicillin-clavulanic acid and ceftriaxone (17% each) (Figure 37). In 2014, no resistance was observed to azithromycin or ciprofloxacin, two antimicrobials used in human medicine for treating severe and invasive salmonellosis.

Among retail chicken meat and slaughter house chicken, the most common Salmonella serotype associated with resistance to third-generation cephalosporin was S. Heidelberg. Further information on third-generation cephalosporin resistance in non-typhoidal Salmonella is provided in the section entitled Integration of AMU and AMR Across the Food chain.

Swine

Recovery of Salmonella spp. among swine samples obtained from farms (grower-finisher), slaughter and retailers has decreased over time from 38% observed in 2004 to 16% in 2014 (Table 13). The majority of the Salmonella isolates were identified as S. Typhimurium, and unlike chicken products, S. Enteritidis, S. Heidelberg and S. Kentucky were rarely identified among samples tested.

In 2014, 33% of all Salmonella isolates recovered from swine samples along the food chain were susceptible to all antimicrobials tested and 44% were identified as multiclass-resistant. The majority of the isolates, 62%, were resistant to tetracycline, followed by sulfisoxazole (44%) and streptomycin (42%). No resistance has been observed to nalidixic acid since the beginning of the surveillance period. Resistance to ampicillin, streptomycin and tetracycline increased between 2004 and 2014 (Figure 38).

Antimicrobial use in Canada

Antimicrobial use in humans

In Canada, antimicrobial use in humans has remained relatively stable over the last 8 years (Figure 39), with only slight changes observed in the total antimicrobials purchased by hospitals (4% decrease) and no changes within the total antimicrobials dispensed by community pharmacies in 2014 compared to 2013. Antimicrobial dispensing in the community continues to account for the majority of all antimicrobial use (AMU) (93%), with a total of 17.8 defined daily doses (DDDs) dispensed per 1,000 inhabitants per day (DIDs) compared to 1.4 DIDs purchased by hospitals.

This accounts for a total of 240,939 kilograms of active ingredients and $786 million spent on antimicrobials in 2014. Changes in the levels of use for individual antimicrobials were observed, with different trends noted in hospital and community settings.

Community utilization

Pharmacy dispensations

On average, 65% of Canadians filled a prescription for an antimicrobial for bacterial infections in 2014 (Figure 40) resulting in a total of 23 million prescriptions dispensed. Prescriptions for amoxicillin represented the largest proportion of all antimicrobials dispensed (26%), followed by azithromycin (9%) and ciprofloxacin (8%). Total antimicrobial use in the community at the national level has remained stable between 2013 and 2014 as measured by DDDs per inhabitant, prescriptions per inhabitant and DDDs per prescription (Table 14). However, between 2007 and 2014, there has been a slight increase in DDDs per prescription and a decrease in the prescriptions per inhabitant.

The majority of antimicrobials used in the Canadian outpatient population in 2014 were drugs for oral administration – at the national level, more than 213 prescriptions and 237 DDDs of oral antimicrobials were dispensed for every parenteral antimicrobial prescription.

Interestingly prescribing rates among children (≤ 14 years old) have decreased 8% in 2014 compared to 2010 and 1% compared to 2013 (Figure 41). In 2014, on average, 63% of children between the ages of 0 and 14 years received an antimicrobial. Fifty-four percent of these prescriptions were for amoxicillin, followed by azithromycin (10%) and clarithromycin (9%). Compared to 2010, amoxicillin and enzyme inhibitor dispensing rates in 2014 increased 27%, while rates for amoxicillin alone increased 8% compared to 2010.

In 2014, 58% of adults (15-59 years old) received an antimicrobial prescription compared to 86% of seniors (60 years and older). Although declines in prescribing rates have been observed between 2010 and 2014 in younger age groups (8% in children and 3% in adults), these declines have not been observed among seniors. Additionally, adults  received more doses per prescription than seniors (approximately 11 days of treatment per prescription, compared to 10 days of treatment for seniors).

Variation in antimicrobial use was observed across Canadian provinces and territories, with Newfoundland and Labrador displaying higher DDDs per inhabitant. In 2014, Newfoundland and Labrador had more than 33% higher DDDs per inhabitant than Saskatchewan (next highest) (Figure 42).

Provincial/territorial variation in the total antimicrobials dispensed at the individual drug level was particularly pronounced for amoxicillin in 2014 (with a high of 3.4 DDDs per inhabitant in Newfoundland and Labrador and a low of 1.3 DDDs per inhabitant in Quebec) (Table 15). Amoxicillin and clarithromycin were the only two drugs identified among the top five antimicrobials dispensed with higher DDDs per inhabitant in every province and territory. Ciprofloxacin was among the top five antimicrobials dispensed in British Columbia (#5), Quebec (#4) and Newfoundland and Labrador (#3).

Prescribing practices by specialization

Information on the specialization of the professional providing a prescription was available in the dispensing data from community pharmacies for 2010 to 2014. A total of 31 different medical and non-medical specializations were identified with an additional category ‘all other speciality’^{Footnote 32}. To facilitate the review of the data, these specializations were further categorized into eight broader groupings^{Footnote 33} (family physicians and general practitioners; dermatologists; diagnostics; emergency medicine; pediatrics, medicine; surgery; and all other specialities).

Sixty-six percent of all prescriptions dispensed by community pharmacies were prescribed by “community practitioners” (family physicians and general practitioners), followed by specialties included in the “all other specialties” (21%) and “medicine” (5%). The antimicrobials most commonly prescribed by “community practitioners” were amoxicillin, azithromycin and ciprofloxacin. While amoxicillin was the top antimicrobial prescribed by practitioners within the “all other specialties” and “medicine” categories, clindamycin and penicillin v rounded off the top three drugs prescribed by “all other specialties” and azithromycin and sulfamethoxazole by “medicine” professionals.

Prescriptions provided by “community practitioners”, “all other specialties” and “medicine” professionals represented the largest number of DDDs per 1,000 inhabitants in 2014 (Figure 43). While the total amounts of antimicrobials prescribed by the different specializations remained relatively stable over time, compared to 2010, the largest decrease in 2014 for the total antimicrobials prescribed was observed by professionals in the “diagnostic” (21% decrease) category. However, an increase was observed in total antimicrobials dispensed for prescriptions in the “medicine” category, increasing 26% in 2014 compared to 2010. These increases were largely due to higher levels dispensed over the five-year period for amoxicillin, azithromycin, cefadroxil, cefazolin, nitrofurantoin and tetracycline.

When looking at the total amount of antimicrobials provided per prescription, dermatologists had the highest levels with 26 DDDs per prescription, followed by “all other specialties” (19 DDDs per prescription) and “medicine” (10 DDDs per prescription) (Figure 44). The larger levels of antimicrobials provided by dermatologists per prescription is consistent with longer treatments provided for skin conditions such as acne. The main drugs prescribed by this specialization included sulfadiazine, doxycycline, tetracycline and trimethoprim products. Prescribing practices have remained stable over the five-year period, with no large changes observed in the total antimicrobials per prescription dispensed by any of the prescriber specializations.

Antimicrobial recommendations by diagnoses

In 2014, a total of 297 million medical diagnoses were made by community-level practitioners resulting in a little over 23 million antimicrobial recommendations. Antimicrobial recommendations are not necessarily tied to a prescription as the patient may not have agreed with receiving an antimicrobial prescription, may not have filled out a prescription following physician orders to wait a period of time, may have chosen not to fill the prescription, may have received a sample of medication, or was a continuation of a previous prescription renewal.

Eight percent of all diagnoses resulted in an antimicrobial being recommended for treatment in 2014. Forty percent of all antimicrobial recommendations were provided for treating respiratory infections, of which 38% were for treating upper respiratory tract infections, followed by acute bronchitis (21%) and acute sinusitis (14%). Overall, 82% of all diagnoses of acute sinusitis resulted in an antimicrobial being recommended followed by acute bronchitis (77%) and pneumonia (74%) (Figure 45).

Children (0 to nine years of age) had the highest overall antimicrobial recommendation rate with nearly eight out of 10 children having been recommended an antimicrobial for treatment in 2014. As a result, this age group had the highest proportion of diagnoses resulting in an antimicrobial recommendation (11%). The majority of the recommendations were provided for the treatment of diseases of the ear (42%), followed by upper respiratory tract infections (24%) and skin and soft tissue infections (8%). Amoxicillin was the predominant antimicrobial recommended for the treatment of diseases of the ear (73% of all recommendations) and upper respiratory tract infections (77%).

While people between 10 and 19 years of age had the lowest recommendation rate (0.51 per person) in 2014, they had the second highest proportion of diagnoses resulting in an antimicrobial recommendation (10%). The majority of the recommendations were provided for treating upper respiratory tract infections (32%), diseases of the ear (12%) and acne (8%). Similar to what was observed in children, amoxicillin was the main antimicrobial recommended for treating diseases of the ear (61%) and upper respiratory tract infections (51%).

Seven out of 10 people in the elderly age group received an antimicrobial recommendation in 2014, mainly for the treatment of lower urinary tract infections or cystitis (16%), acute bronchitis (11%) and skin and soft tissue infections (10%). In 2014, 39% percent of all recommendations provided for treating skin and soft tissue infections were for cephalexin. For treating acute bronchitis, a similar proportion of recommendations were made for azithromycin (27%) and clarithromycin (25%). Similarly, ciprofloxacin (41%) and nitrofurantoin (41%) were recommended for treating lower urinary tract infections or cystitis.

On average, the percentage of diagnoses for which an antimicrobial was recommended remained stable over time, while the antimicrobial recommendation rates per population decreased (Figure 46).

Hospital purchasing and use

In 2014, 38,340 kilograms of antimicrobials were purchased by hospitals across Canada at a cost of $104 million. When adjusted for the number of hospital discharges that occurred in 2014, this amounts to approximately six DDDs of antimicrobial purchased per discharge, a rate that has remained fairly stable over the 2002 to 2014 period of surveillance (Figure 47).

Ciprofloxacin was the antimicrobial most commonly purchased in 2014 (10% of all antimicrobial DDDs purchased), followed by amoxicillin, azithromycin, ceftriaxone and doxycycline (7% each) (Table 10). Although the purchasing of amoxicillin and ciprofloxacin remained stable between 2002 and 2014, ceftriaxone and doxycycline had the highest increase during the same period. Ceftriaxone purchases increased from 79 DDDs per 1,000 discharges in 2002 to 440 DDDs per 1,000 discharges in 2014 and doxycycline increased from 124 to 417 DDDs per 1,000 discharges during the same time period.

Similar to the variation observed in the community dispensing data, large variations in hospital purchasing of antimicrobials were present in 2014 among the provinces. Manitoba had the largest number of DDDs purchased per discharge than any other province since 2003, 49% higher than Nova Scotia, the province with the next highest number.

Figure 47: Defined daily doses per patient discharge for antimicrobials purchased by hospitals in Canada, by province, 2014

Large differences in the purchasing of specific antimicrobials were observed in Manitoba compared to the other provinces (Table 17), particularly amoxicillin (2,302 DDDs per 1,000 discharges compared to 225 DDDs per 1,000 discharges in Nova Scotia), cefoxitin (1,497 DDDs per 1,000 discharges compared to 60 DDDs per 1,000 discharges in Newfoundland and Prince Edward Island) and cephalexin (877 DDDs per 1,000 discharges compared to 93 DDDs per 1,000 discharges in Quebec). Similarly, Nova Scotia purchased larger amounts of imipenem and cilastatin (54 DDDs per 1,000 discharges) compared to Quebec (17 DDDs per 1,000 discharges) which had the second highest levels and Saskatchewan (0.01 DDDs per 1,000 discharges) representing the lowest levels in 2014.

Unlike the community dispensing data where commonalities were observed among the top five antimicrobials dispensed across the provinces, large variations were observed in the antimicrobials purchased by hospitals. Ceftriaxone was the only antimicrobial identified among the top five antimicrobials purchased by all provincial hospitals with the exception of Quebec. More specifically, the top antimicrobial purchased in each province varied with ceftriaxone identified as the top antimicrobial in Alberta, doxycycline in British Columbia, amoxicillin in Manitoba, Newfoundland and Labrador and Prince Edward Island, azithromycin in Saskatchewan, cefazolin in New Brunswick and ciprofloxacin in Ontario, Quebec and Nova Scotia. Nitrofurantoin was identified among the top 10 antimicrobials purchased by hospitals in British Columbia in 2014, but was not identified within the top 10 antimicrobials for any other province. The reasons for the differences in antimicrobial use between provinces are not well understood, but are likely (at least in part) due to different drug formularies and different treatment protocols and variation in sub specialties care units such as Transplantation, Burn and Oncology.

International perspective

The European Surveillance of Antimicrobial Consumption Network collects information from European member countries to estimate the total antimicrobial consumption in both the community and the hospital sectors.

In 2014, a total of 30 European countries provided information to ESAC-Net on antimicrobials consumed in their community. When these data were compared with the 2014 Canadian outpatient antimicrobial consumption rate, Canada (17.8 DDDs per 1,000 persons per day) ranked 12th lowest out of 31 countries by increasing level of AMU, with almost half the level of use reported by Greece (country with highest use, 34 DDDs per 1, 000 persons per day) (Figure 48)^{Footnote bq}.

On the other hand, 23 European countries provided information to ESAC-NET on total antimicrobials purchased by their hospitals in 2014. Compared to these data, Canada (1.4 DDDs per 1,000 discharges per day) ranked third lowest out of 24 countries classified by increasing levels of total AMU in hospitals, with nearly half the level of antimicrobials purchased by hospitals in the United Kingdom (2.6 DDDs per 1,000 persons per day) (Figure 49)^{Footnote bq}.

In the United States, in 2011, healthcare providers prescribed 262.5 million courses of antibiotics (842 prescriptions per 1,000 persons) in community settings. Acute respiratory infections result in the vast majority of inappropriate antibiotic prescriptions in United States outpatient clinics. In 2010, 71% of all outpatient visits for acute bronchitis resulted in an antibiotic prescription. On the other hand, between the mid 1990’s and 2008, outpatient antibiotic prescribing rates for children (≤ 14 yrs.) with acute respiratory tract infections decreased by 11%^{Footnote br}.

As such, inappropriate antibiotic use is also seen in United States hospitals, where in 2010, 55.7% of patients discharged from 323 United States hospitals received antibiotics during their hospitalization and antibiotic use could have been improved in almost 37.2% of the reviewed prescription scenarios^{Footnote br}.

Figure 48: Outpatient antimicrobial use (DDD per 1,000 persons per day) reported in Canada and in 30 European countries

Figure 49: Hospital antimicrobial purchases (DDD per 1,000 persons per day) reported in Canada and in 23 European Countries

Antimicrobial use in animals

Total antimicrobials distributed for sale for use in animals

Information on antimicrobials distributed for sale for use in animals has been voluntarily provided by the Canadian Animal Health Institute (CAHI) since 2006. These data represent quantities of antimicrobials distributed for sale by member companies and do not include quantities of antimicrobials imported for own use or as active pharmaceutical ingredients used in further compounding.

In 2014, 1.5 million kg of antimicrobial active ingredients were distributed for use in animals. This number is 5% higher than in 2013 but 12% lower than reported in 2006. Over the past five years (2010 to 2014), there has been a 1% increase. In Canada, 99% of the antimicrobials distributed in 2014 were intended for use in food-production animals and less than 1% was intended for use in companion animals. Additionally, the majority (73%) of antimicrobials distributed were in the same classes as those used in human medicine. Inappropriate antimicrobial use in food-producing animals is a public health concern as it contributes to emergence of resistant bacteria in animals that can be transmitted to humans through food supply^{Footnote 35}.

As in previous years, the predominant classes of antimicrobials distributed for use in animals in descending order were the tetracyclines, ionophores, β-lactams, other antimicrobials^{Footnote 36} and the macrolides. Fluoroquinolones are classified as “very high importance to human medicine^{Footnote 37} by Health Canada’s Veterinary Drugs Directorate. Fluoroquinolones are licensed for use in certain animal species in Canada and have warnings on their labels recommending against extra-label use due to antimicrobial resistance concerns. The overall quantity of fluoroquinolones distributed for use in animals increased by 14% between 2013 and 2014, though there has been a 40% increase since 2010 (likely due in part to approval of a new fluoroquinolone indication for use).

Third-generation cephalosporins are also of very high importance to human medicine, are licensed for use in some animal species in Canada and bear the same antimicrobial resistance warning statement on their labels. CAHI data show a decline of 60% in the quantity of cephalosporins distributed for use in animals from 2011 to 2014^{Footnote 38}.

There were provincial differences between the quantities and types of antimicrobials distributed and differences within provinces over time such as increases in antimicrobials distributed for sale between 2013 and 2014 were observed for British Columbia, Alberta, Manitoba and Ontario; the most notable increases occurred in Ontario and Alberta. Quebec and the Atlantic provinces all had decreases in antimicrobials distributed for sale between 2013 and 2014. These values do not account for changes in underlying populations or disease pressures. It should be noted that interprovincial distribution of antimicrobials can occur after surveillance data are captured.

Figure 50: Quantity of antimicrobials (kg) distributed for sale for use in animals, by province, 2011 to 2014

Data source: Canadian Animal Health Institute.
Values do not include own use imports or active pharmaceutical ingredients used in compounding.
There may be subsequent distribution of antimicrobials across provincial borders after being distributed to the veterinary clinics.
This figure does not account for provincial differences in numbers or types of animals or disease pressures.