Report on Antimicrobial Use and Resistance in Humans in 2012

Report on Antimicrobial Use and Resistance in Humans in 2012

Acknowledgements This report could not have been compiled without antimicrobial resistance data provided by diagnostic and reference laboratories within Scotland and the prescribing team at Information Service Division (ISD) who assisted with analysis of antimicrobial use data. We also extend thanks to the Scottish Microbiology and Virology Network for advice on the development of this programme and their ongoing input into national surveillance of antimicrobial resistance. Health Protection Scotland (HPS) and Information Services Division (ISD) are divisions of NHS National Services Scotland. HPS website: http://www.hps.scot.nhs.uk ISD website: http://www.isdscotland.org/ Published by Health Protection Scotland and Information Services Division First published January 2014 Health Protection Scotland and Information Services Division 2014 ISBN: 978-1-873772-45-4 Reference this document as: Health Protection Scotland and Information Services Division. Report on Antimicrobial Use and Resistance in Humans in 2012. Health Protection Scotland and Information Services Division. 2014 [Report] HPS and ISD have made every effort to trace holders of copyright in original material and to seek permission for its use in this document. Should copyrighted material have been inadvertently used without appropriate attribution or permission, the copyright holders are asked to contact HPS or ISD so that suitable acknowledgement can be made at the first opportunity. HPS and ISD consent to the photocopying of this document for professional use. All other proposals for reproduction of large extracts should be addressed to either: Health Protection Scotland NHS National Services Scotland Meridian Court 5 Cadogan Street Glasgow G2 6QE Tel: +44 (0) 141 300 1100 Email: nss.hpsenquiries@nhs.net Designed and typeset by: Graphics Team, Health Protection Scotland Healthcare Information Group Information Services Division Gyle Square 1, South Gyle Crescent Edinburgh EH12 9EB Tel: +44 (0) 131 275 7777 Email: nss.isdprescribing@nhs.net

Table of Contents List of abbreviations and acronyms Executive Summary 1. Introduction 1 2. Methods 2 3. Clinical priorities 3 Carbapenem use and the evolving threat from global spread of carbapenemaseproducing Enterobacteriaceae 3 Antimicrobial treatment of urinary tract infection and development of resistance in urinary isolates 7 Epidemiology of Clostridium difficile Infection (CDI) and impact of antimicrobial stewardship 10 Prevention of Streptococcus pneumoniae by vaccination and impact on resistance patterns 12 Daptomycin use and potential for development of resistance 14 4. Development of health information systems to support improved prevention and treatment of infection 16 Development of the NHSScotland Infection Intelligence Platform 16 The Electronic Communication of Surveillance in Scotland (ECOSS) web system 17 5. Results 19 5.1. Antimicrobial use in primary care 19 Total use of antibacterials 19 Frequency of use of antibacterials in primary care 21 Choice of antibacterial in primary care 24 Antibacterial use by age-group in primary care 26 Antibacterial use in dentistry 28 Non-medical prescribing in primary care 29 5.2. Antimicrobial use in secondary care (hospitals) 30 Total use of antibacterials in hospitals 30 Route of administration of antibacterials in hospital patients 31 Choice of antibacterial in hospitals 32 Alert antibacterials in hospitals 35 5.3. Resistance 36 Gram-negative bacteraemia 36 ESBL-producers in Gram-negative bacteraemias 36 Escherichia coli 37 Aminopenicillins (E. coli) 38 Cephalosporins (E. coli) 38 vii ix Report on Antimicrobial Use and Resistance in Humans in 2012

Carbapenems (E. coli) 39 Aminoglycosides (E. coli) 39 Fluoroquinolones (E. coli) 39 Trimethroprim (E. coli) 39 Aztreonam (E. coli) 39 Combined resistance to third-generation cephalosporins, fluoroquinolones and aminoglycosides (E. coli) 39 Klebsiella pneumoniae 40 Cephalosporins (K. pneumoniae) 41 Carbapenems (K. pneumoniae) 41 Aminoglycosides (K. pneumoniae) 42 Fluoroquinolones (K. pneumoniae) 42 Trimethroprim (K. pneumoniae) 42 Combined resistance to third-generation cephalosporins, fluoroquinolones and aminoglycosides (K. pneumoniae) 42 Pseudomonas aeruginosa 42 Cephalosporins (P. aeruginosa) 44 Carbapenems (P. aeruginosa) 44 Aminoglycosides (P. aeruginosa) 44 Fluoroquinolones (P. aeruginosa) 44 Piperacillin-tazobactam (P. aeruginosa) 44 Combined resistance to cephalosporins, fluoroquinolones and aminoglycosides (P. aeruginosa) 44 Acinetobacter baumannii 45 Cephalosporins (A. baumannii) 45 Carbapenems (A. baumannii) 45 Aminoglycosides (A. baumannii) 45 Piperacillin-tazobactam (A. baumannii) 45 Haemophilus influenzae 46 Neisseria meningitidis 47 Gram-positive bacteraemia 47 Staphylococcus aureus 49 Meticillin sensitive S. aureus (MSSA) 50 Meticillin resistant S. aureus (MRSA) 51 Streptococcus pneumoniae 52 Enterococcus spp. 53 6. Recommendations 55 7. Update on previous recommendations 56 References 59 Report on Antimicrobial Use and Resistance in Humans in 2012

List of Tables Table 1: Number of organisms included in the surveillance of AMR of urinary tract infections from January 2012 - September 2013 8 Table 2: Resistance in E. coli urinary isolates from January 2012 - September 2012 and January 2013 - September 2013 9 Table 3: Use of antibacterials in primary care (excluding dental) in UK countries from 2009-2012, items/1 000/day, percent change previous year 20 Table 4: NHSScotland, number of individuals dispensed the same antibacterial item 6 times over the course of 2012 (CHI completeness 89.6%) 23 Table 5a: NHSScotland: use of antibacterials in primary care (excluding dental), items/100 000/day (percent annual change), by group 2009-2012 25 Table 5b: NHSScotland: use of antibacterials in primary care (excluding dental), DDD/100 000/day (percent annual change) by group 2009-2012 25 Table 6: Use of antibacterials associated with CDI in primary care (excluding dental) in UK countries from 2009-2012, items/100 000/day, percent change previous year 26 Table 7a: NHSScotland: use of antibacterials in primary care (excluding dental) by age group in 2012, items/100 000/day, percent change from 2011 (CHI capture presented by ATC classification) 27 Table 7b: NHSScotland: use of antibacterials in primary care (excluding dental) by age group in 2012, items/100 000/day, percent change from 2011 (CHI capture presented by ATC classification) 28 Table 8: NHSScotland: use of antibacterials in secondary care 2009-2012, DDD/100 000/day, (percent change previous year) in 11 NHS boards covering 84% population 34 Table 9: Gram-negative bacteraemias in 2009-2012; number of cases according to adapted EARS-Net definition and percentage of ESBL-producers 36 Table 10: Percentage resistance (and number of isolates tested) in E. coli in 2009-2012 37 Table 11: Percentage resistance (and number of isolates tested) in K. pneumoniae in 2009-2012 40 Table 12: Percentage resistance (and number of isolates tested) in P. aeruginosa bacteraemias in 2009-2012 43 Table 13: Percentage resistance (and number of isolates tested) in A. baumannii bacteraemias in 2009-2012 45 Table 14: Percentage resistance in H. influenzae 2009-2012 46 Table 15: Percentage resistance in N. meningitidis 2009-2012 47 Table 16: Gram-positive bacteraemias in 2009-2012; number of cases as per adapted EARS-Net definition 47 Table 17a: Percentage resistance in MSSA bacteraemias 2009-2012. All isolates were susceptibility tested 48 Report on Antimicrobial Use and Resistance in Humans in 2012 iii

Table 17b: Percentage resistance in MRSA bacteraemias 2009-2012. All isolates were susceptibility tested 48 Table 18: Percentage resistance in S. pneumoniae bacteraemias 2009-2012. All isolates were susceptibility tested 49 Table 19a: E.faecalis - vancomycin percentage resistance in bacteraemias 2009-2012 49 Table 19b: E. faecium - vancomycin percentage resistance in bacteraemias 2009-2012 49 iv Report on Antimicrobial Use and Resistance in Humans in 2012

List of Figures Figure 1: NHSScotland: use of carbapenems in secondary care, DDD/100 000/day, 2009-2012, (11 NHS boards covering 84% of the population) 3 Figure 2: Carbapenemase producers reported in Scotland reported by AMRHAI (PHE) 4 Figure 3: Epidemiological stages of CPE in Europe, 2013 (from Glasner et al, 2013) 6 Figure 4: Overall quarterly CDI incidence rates for Scotland (per 100 000 occupied bed days) in patients 65 years and 15-64 years for the period October 2006 to December 2012 11 Figure 5: Distribution of the most common Scottish S. pneumoniae serotypes (%), 2010, 2011 and 2012 13 Figure 6: NHSScotland: use of daptomycin in secondary care, DDD/100 000/day, 2009-2012, (11 NHS boards covering 84% of the population) 14 Figure 7: NHSScotland: use of antibacterials in primary care, items/1 000/day and DDD/1 000/day 1995-2012 19 Figure 8: NHSScotland: use of antibacterials by GP practices: items/1 000/day, October to December 2012 20 Figure 9: NHSScotland: number of individuals prescribed antibacterials by number of items dispensed during that year, (excludes dental prescriptions and based on 89.6% prescriptions with CHI number for antibacterials) 21 Figure 10: NHSScotland: number of individuals prescribed antibacterials in 2012 by age (excluding dental) and based on 89.6% prescriptions with CHI number for antibacterials 22 Figure 11: NHSScotland: use of antibacterials in primary care by age and gender, items/1 000/day 2012. CHI completeness 89.6% 24 Figure 12: NHSScotland: high-risk, recommended and other antibacterials as a proportion of all nurse antibacterial prescribing in terms of items, 2007-2012 29 Figure 13: NHSScotland: use of antibacterials in secondary care, DDD/100 000/day, 2009-2012, (11 NHS boards covering 84% of the population) 31 Figure 14: NHSScotland: parenteral use as a proportion of all use (DDD) for selected and all antibacterials, 2009-2012, (11 NHS boards covering 84% of the population) 32 Figure 15: NHSScotland: use of alert antibacterials, DDD/100 000/day, 2009-2012 in 11 NHS boards (covering 84% of population) 35 Figure 16: Antimicrobial resistance (with 95% confidence intervals) in E.coli isolated from blood cultures in 2009 (n=3 486), 2010 (n=3 602), 2011 (3 839) and 2012 (3 924) 38 Figure 17: Antimicrobial resistance (with 95% confidence intervals) in K. pneumoniae isolated from blood cultures in 2009 (n=672), 2010 (n=715), 2011 (n=697) and 2012 (n=718) 41 Figure 18: Antimicrobial resistance (with 95% confidence intervals) in P. aeruginosa isolated from blood cultures in 2009 (n=269), 2010 (n=295), 2011 (242) and 2012 (234) 43 Report on Antimicrobial Use and Resistance in Humans in 2012 v

Figure 19: Antimicrobial resistance (with 95% confidence intervals) in meticillin-sensitive Staphylococcus aureus (MSSA) isolated from blood cultures in 2009 (n=1 348), 2010 (n=1 317) 2011 (n=1 258) and 2012 (n=1 187) 50 Figure 20: Antimicrobial resistance (with 95% confidence intervals) in meticillin-resistant Staphylococcus aureus (MRSA) isolated from blood cultures in 2009 (n=478), 2010 (n=310) 2011 (n=194) and 2012 (n=173) 51 Figure 21: Antimicrobial resistance (with 95% confidence intervals) in S. pneumoniae isolated from blood cultures in 2009 (n=579), 2010 (n=479) and 2011 (n=438) and 2012 (n=399) 52 Figure 22: Vancomycin resistance (with 95% confidence intervals) in E. faecalis isolated from blood cultures in 2009 (n=447), 2010 (n=397), 2011 (n=363) and 2012 (n=419) 53 Figure 23: Vancomycin resistance (with 95% confidence intervals) in E. faecium isolated from blood cultures in 2009 (n=255), 2010 (n=230), 2011 (n=214) and 2012 (n=250) 54 vi Report on Antimicrobial Use and Resistance in Humans in 2012

List of abbreviations and acronyms AMR AMT AMRHAI AST ATC BNF CDI CHI CLSI CPE CSO DDD EARS-Net ECDC ECOSS ESBL Antimicrobial resistance Antimicrobial management team Antimicrobial Resistance and Healthcare Associated Infections Reference Unit (HPA) Antimicrobial susceptibility testing Anatomical Therapeutic Chemical British National Formulary Clostridium difficile infection Community Health Index Clinical and Laboratory Standards Institute Carbapenemase producing Enterobacteriaceae Chief Scientist Office Defined daily dose European Antimicrobial Resistance Surveillance Network (formerly EARSS) European Centre for Disease Prevention and Control Electronic Communication of Surveillance in Scotland Extended spectrum beta-lactamase EuSCACPE European Survey on carbapenemase producing Enterobacteriaceae EWRS HMUD HPS ISD KPC MIC MRSA MSSA Early Warning and Response System Hospital Medicines Utilisation Database Health Protection Scotland (NHS National Services Scotland) Information Services Division (NHS National Services Scotland) Klebsiella pneumoniae carbapenemase Minimum inhibitory concentration Meticillin resistant Staphylococcus aureus Meticillin sensitive Staphylococcus aureus NDM-1 New Delhi metallo beta-lactamase (type 1) NHS NSS OCBD PCR PIS National Health Service National Services Scotland Occupied bed days Polymerase chain reaction Prescribing information system Report on Antimicrobial Use and Resistance in Humans in 2012 vii

PPS PRISMS SAPG Point Prevalence Survey Prescribing Information System for Scotland Scottish Antimicrobial Prescribing Group ScotMARAP Scottish Management of Antimicrobial Resistance Action Plan SMC SHLMPRL SMRSARL SSSCDRL UK VIM WHO Scottish Medicines Consortium Scottish Haemophilus, Legionella, Meningococcus and Pneumococcus Reference Laboratory Scottish MRSA Reference Laboratory Scottish Salmonella, Shigella and Clostridium difficile Reference Laboratory United Kingdom Verona integron-encoded metallo-beta-lactamase World Health Organisation viii Report on Antimicrobial Use and Resistance in Humans in 2012

Executive Summary This is the fifth annual report published by Health Protection Scotland (HPS)/ Information Services Division (ISD), part of the Public Health and Intelligence (PHI) strategic business unit of NHS National Services Scotland (NSS), that combines information on antimicrobial use and resistance in humans in Scotland. The information presented here covers up to the end of 2012. Since 2009, there has been a 23.1% increase in the use of carbapenems in Scottish hospitals. Resistance to carbapenems among Gram-negatives remains rare although sporadic reports of confirmed carbapenemase producers are received. There has been a continued increase in the use of trimethoprim and nitrofurantoin since 2009, suggestive of increased compliance with prescribing policies for urinary tract infection. However, 11 538 persons received trimethoprim or nitrofurantoin on six or more occasions, suggesting prophylactic use, although there is limited evidence for this practice. Resistance to trimethoprim increased from 38.0% to 40.3% and nitrofurantoin from 3.9% to 4.7% from 2011 to 2012. In order to reduce this unnecessary use of these antibacterials regular reviews of all patients in primary care receiving antibacterial prophylaxis are required. In 2012, the use of antibacterials in primary care was 3.3% higher than in 2011, corresponding to an increase of 140 438 prescription items. This is the second consecutive year an increase has been observed. Overall 33.0% of the Scottish population had at least one antibacterial item dispensed in 2012, while 8.3% had 3 items and 2.1% had 6 items dispensed. Since 2010, there has been an upward trend in the proportion of the Scottish population who received antibacterials. However, there was continued progress in 2012 toward reduction in the use of broadspectrum antibacterials associated with Clostridium difficile infection (CDI) in primary care; cephalosporins 17.6%; combination penicillins (mainly co-amoxiclav) 8.3% and fluoroquinolones 8.0%. In 2012, the total use of antibacterials in hospitals was 6.2% higher than in 2011 continuing the upward trend since 2009. Work to analyse reasons for this increase will be a key priority in 2014. In 2012, the use of broad-spectrum antibacterials associated with CDI increased by 7.2% in hospitals, largely accounted for by increases in co-amoxiclav use (12.2%) and clindamycin use (7.9%). The use of cephalosporins and fluoroquinolones were unchanged from 2011, but there were increases in the use of glycopeptides (10.0%) and piperacillin in combination with tazobactam (9.2%) in 2012. Following substantial reductions in CDI incidence from 2008 to 2010, trends have flattened since 2011. There is a need to drive further reductions in CDI rates. Report on Antimicrobial Use and Resistance in Humans in 2012 ix

In 2012, Escherichia coli (3 925 cases) and Klebsiella pneumoniae (718 cases) continued to be the most frequent causes of Gram-negative bacteraemia, and a continuing upward trend in burden of disease was observed for both organisms. The number of cases of Pseudomonas aeruginosa bacteraemia remained stable (234 cases). The number of cases of Acinetobacter baumannii (23 cases) continued to decrease in 2012. Resistance among E. coli causing bacteraemia has remained stable for the last four years and significant decreases in resistance, comparing the year 2011 to 2012, were reported for ampicillin, co-amoxiclav, cefotaxime, ceftazidime, ciprofloxacin and piperacillin/ tazobactam. In 2012, 6.6% of the E. coli bacteraemia isolates were ESBL producers. Resistance among K. pneumoniae bacteraemias has decreased since 2009 for most agents; with resistance to the third-generation cephalosporins (ceftriaxone, cefotaxime and ceftazidime) within the range of 6.5% to 8.5% and gentamicin 7.3% in 2012. In 2012, 6.4% of K. pneumoniae bacteraemia isolates were ESBL producers. Although resistance among the Scottish Gram-negative isolates have shown stable or decreasing trends since 2009 resistance to nearly all clinically important classes of antibacterials, including aminopenicillins, cephalosporins, fluoroquinolones and aminoglycosides have continuously been reported in Gram-negative bacteraemias in the period 2009-2012. In 2012, the lowest number of cases of Staphylococcus aureus (1 360 cases) was reported since 2009 (1 826 cases), which was mainly caused by reductions in meticillin resistant S. aureus (MRSA) while a concomitant reduction has not been seen in numbers of meticillin sensitive S. aureus (MSSA). The proportion of MRSA among S. aureus bacteraemias decreased further in 2012 to 12.7%. Whilst a reduction in vancomycin resistance was observed among the Scottish isolates Enterococcus faecium in 2012 (24.4%), this was above the proportion reported for the whole of the UK (13.3%) and is the third highest percentage reported in Europe. The governments of the four UK nations have collaborated to produce the UK wide five year antimicrobial resistance strategy 2013-2018 [https://www.gov.uk/government/ publications/uk-5-year-antimicrobial-resistance-strategy-2013-to-2018]. The surveillance of antimicrobial use and resistance will be aligned across the UK and the Scottish Antimicrobial Prescribing Group programme will be developed to address specific clinical priority areas outlined in this strategy. x Report on Antimicrobial Use and Resistance in Humans in 2012

1. Introduction Antimicrobial resistance (AMR) continues to pose a serious public health threat globally. The loss of effective antimicrobials undermines our ability to fight infectious diseases and manage the infectious complications common in vulnerable patients undergoing chemotherapy for cancer, dialysis, and surgery, especially organ transplantation, for which the ability to treat secondary infections is crucial. 1 The problem is further exacerbated by the fact that a new infectious disease has been discovered almost every year over the past 30 years, while very few new antimicrobials have been developed. In 2013, antimicrobial resistance was added to UK national risk register to ensure it receives adequate attention of policy makers in the UK. This is the fifth annual report on antimicrobial use and resistance in humans which underpins the work of the Scottish Antimicrobial Prescribing Group (SAPG) national antimicrobial stewardship programme. The national programme, established in 2008, coordinates Scotland-wide activities aimed at driving improvement in how we use antimicrobials to treat and prevent infection in hospitals and the community, monitor the evolving threat from resistant bacteria and develop best practice in management of infection, while linking with other national programmes aimed at preventing and controlling healthcare associated infections and antimicrobial resistance. This report presents detailed information on antimicrobial use and resistance in Scotland covering the period to the end of 2012 and provides recommendations on interventions and national healthcare improvement activities aimed at stemming the tide of further resistance development. The report is intended to support National Health Service (NHS) boards and national health organisations in their long-term strategic planning, implementation and evaluation of interventions and quality improvement activities in relation to antimicrobial prescribing, infection management, and infection prevention and control. Reducing healthcare associated infection and containing antimicrobial resistance remain key priorities for NHSScotland in order to deliver high quality, safe and person centred healthcare. 2;3 The initial priority for antimicrobial stewardship in Scotland, aimed at reducing the use of agents with high risk of Clostridium difficile infection (CDI), has been effective in improving the quality of antimicrobial prescribing in primary care and reducing the incidence of CDI in Scotland. The focus has now moved to reducing unnecessary antimicrobial use which is aimed at reducing the selection pressure for antimicrobial resistance. A recent European epidemiological survey of carbapenemase producing Enterobacteriaceae (EuSCACPE) reported that a higher state of spread of CPE was observed in 2013 compared with 2010 in more than half of countries participating in this and two previous surveys. 4 The overall trend for the UK was reported to be increasing from sporadic hospital outbreaks in 2010 to regional spread in 2012 and 2013, while Scotland continued to report a sporadic stage. Importantly, under-detection and under-reporting was indicated by respondents in several countries leading to uncertainty about the true epidemiological stage in these countries. These recent changes in the epidemiology of AMR in the UK and across Europe and worldwide, highlights the continued need for action to improve the way we use antimicrobials, detect and monitor threats from emerging resistant microorganisms, including maintaining high awareness, good diagnostic practice and surveillance and containment through infection control practices and screening. Report on Antimicrobial Use and Resistance in Humans in 2012 1

For the first time, since SAPG was established in 2008, a new joint UK AMR strategy for 2013-2018 (agreed between the governments of the four UK nations and involving human and animal health agencies), aimed at slowing the development and spread of antimicrobial resistance, has been developed. 5 In order to further develop and progress the activities within the SAPG antimicrobial stewardship programme, specific clinical priority areas and recommendations underpinning the existing Scottish Action Plan 6 and the new UK strategy will be presented in this report. https://www.gov. uk/government/publications/uk-5-year-antimicrobial-resistance-strategy-2013-to-2018. 2. Methods The methods can be accessed on the ISD website: http://www.isdscotland.org/health-topics/ Prescribing-and-Medicines/National-Medicines-Utilisation-Unit/Antimicrobial-Prescribing-and- Resistance.asp. 2 Report on Antimicrobial Use and Resistance in Humans in 2012

3. Clinical priorities Carbapenem use and the evolving threat from global spread of carbapenemase-producing Enterobacteriaceae Multidrug resistance among Gram-negative organisms continues to be a major threat to public health and patient safety. In particular, the emergence of carbapenemase producing Enterobacteriaceae (CPE) is concerning as this leaves only few options to treat severe infections as these organisms often are resistant to the majority of all available antibacterials. Moreover, CPE s spread rapidly across healthcare systems around the world via clonal dissemination and spread via plasmid transfer both inside and outside healthcare settings. In 2012, the use of carbapenems was 9.1% higher than in 2011 and continues the increasing trend observed in previous years (Figure 1). Meropenem accounted for 95.3% of carbapenem use in 2012. Although carbapenems use accounted for only 1.3% of total antibacterial use in 2012, since 2009 there has been a 23.1% increase in carbapenems use in Scotland. It is speculated that this increase could have been driven partly by initiatives to reduce the use of cephalosporins and fluoroquinolones as part of national approach to reduce CDI. SAPG should consider undertaking an audit to understand whether the use of carbapenems was appropriate and consistent with local prescribing guidance. Figure 1: NHSScotland: use of carbapenems in secondary care, DDD/100 000/day, 2009-2012, (11 NHS boards covering 84% of the population) 6 DDDs / 100 000 pop / day 5 4 3 2 1 0 2009 2010 2011 2012 Year In Scotland, 21 carbapenemase producers were reported from AMRHAI (PHE) in 2013, bringing the total number of reports to 100 since 2003. Figure 2 shows the distribution of phenotypes among carbapenemase producers in Gram-negatives in the period 2003-2013. Reports of carbapenemase producers have been geographically widespread across Scotland since 2009. In 2013, carbapenemase producers were reported in seven mainland NHS boards. To date, 11 out of 14 NHS board areas have now reported at least one carbapenemase producer. Report on Antimicrobial Use and Resistance in Humans in 2012 3

Based on the wide geographical distribution and occurrence of diverse species and phenotypes, it is currently assumed that carbapenemases in Scotland occur sporadically (i.e. there is no spread within hospitals, geographical regions or beyond). However, one event of possible transmission was reported to HPS in 2013, and other single events of transmission may have occurred which have not been identified due to lack of testing and sub-typing. In 2013, KPC-K. pneumoniae was the most frequently reported (7/21) carbapenemase-producer in Scotland. NDM-producers including E. coli, K. pneumoniae and Enterobacter cloacae (complex) were reported on seven occasions in five different Scottish NHS boards. Worldwide, K. pneumoniae KPC is the most widespread carbapenemase producer due to the clonal spread of a specific lineage of K. pneumoniae (ST258). In Europe, KPC producing Enterobacteriaceae also has the widest distribution overall, and is the most frequently reported CPE in the majority of countries. 4 In contrast, NDM-producers are still very rare in Europe, but the UK continues to report more NDM-positive isolates than other countries. Importation via travel from the Indian subcontinent is playing an important role in the spread of NDM in Europe and elsewhere. The fact that NDM is disseminated via plasmid transmission among the human Gram-negative gut flora makes this type of resistance potentially more difficult to control than the clonally spread CPEs. Moreover, it has been estimated that significant proportions of the populations in India and other South Asian countries carry NDM-producers in their gut flora, while many live in environments with inadequate sanitation and no access to potable water, which exacerbate the situation and continuously drive further spread of this type of resistance. 7 Figure 2: Carbapenemase producers reported in Scotland reported by AMRHAI (PHE) 30 25 Number of isolates 20 15 10 5 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Year OXA-48 NDM-1 IMI IMP VIM KPC 4 Report on Antimicrobial Use and Resistance in Humans in 2012

In a recent European survey, the epidemiological stages (ranging from sporadic to endemic) were determined in 39 countries. National experts assessed the stage of spread in their respective countries. Sporadic spread was reported by 22 countries (59%) while regional/national spread was reported by 11 countries (25%). Three countries (Greece, Italy and Malta) reported spread of CPE to be in the endemic stage. Scotland also participated in the self-assessment exercise and reported the spread of CPE to be sporadic. In order to be aligned with ECDC reporting practices, it was agreed to report the Scottish epidemiological status jointly with the UK, which overall was reported as a country with regional spread of CPE (due to having more than one hospital outbreak and regional inter-institutional spread). A number of countries reported that they were uncertain about the spread of CPE due to lack of systematic surveillance (indicated by hatching on the map in Figure 3). Among 31 countries that previously had participated in a similar exercise in 2010, a higher state of spread was reported in 17 (55%) countries. Although detection of CPE may have been improved in this three-year period this finding suggests a very rapid spread of carbapenemases across Europe. The previous report recommended the development of a strategy to preserve the effectiveness of carbapenems for future use, to include development of guidance on prescribing and nonprescribing (infection prevention and control) measures and strengthen local and national surveillance. To support this recommendation, a sub-group of SAPG developed national consensus guidance on treatment of multidrug resistance Gram-negative infections, which propose areas where alternative antibacterials could be considered to reduce the use of carbapenems. This guidance aims to support Antimcrobial Management Teams (AMTs) and infection specialists to develop local prescribing guidance. The recommendations provide pragmatic advice on alternative options to manage MDR Gram-negative infections until more definitive evidenced based guidance becomes available; and suggests the use of a range of alternative agents such as aztreonam, temocillin, pivmecillinam and fosfomycin (unlicensed in the UK) for empirical and targeted treatment of Gram-negative infections. We will monitor the use of carbapenems following publication of this guidance and will update SAPG regularly and highlight changes in the next report. Two larger Scottish NHS boards currently participate in a Europe-wide exercise (under the EuSCAPE programme) aimed at building diagnostic laboratory and surveillance capability and harmonising the approach in all member states. The results of this epidemiological survey, which will be available towards the end of 2014, will inform the further development of surveillance of CPE s in Scotland. Moreover, HPS published a guidance on interim non-prescribing control measures to prevent cross transmission of CPE in acute settings. 8 Report on Antimicrobial Use and Resistance in Humans in 2012 5

Figure 3: Epidemiological stages of CPE in Europe, 2013 (from Glasner et al, 2013) Epidemiological stages Non-visible countries No cases reported Sporadic occurance Single hospital outbreak Sporadic hospital outbreak Regional spread Inter-regional spread Endemic situation Data not available Not participating Uncertain Luxembourg Malta Cyprus Source: Eurosurveillance (February 2013) available online: http://www.eurosurveillance.org/viewarticle. aspx?articleid=20525. 6 Report on Antimicrobial Use and Resistance in Humans in 2012

Antimicrobial treatment of urinary tract infection and development of resistance in urinary isolates Urinary tract infections (UTIs) are among the most commonly encountered infections in primary care, care home and hospitals. Antibacterials for UTI are one of the most common reasons for empirical use of antibacterials across all settings. As such, the treatment and prevention of UTI remains a clinical priority for the antimicrobial stewardship and infection prevention and control clinical communities. A key component of the national approach to antimicrobial stewardship co-ordinated by SAPG has included the development of frameworks to support the development of prescribing guidance in primary and secondary care. Trimethoprim and nitrofurantoin are commonly recommended for the empirical treatment of UTI. 9;10 From 2011 to 2012, there was an increase in use of trimethoprim (and sulphonamides) (4.1%) and nitrofurantoin (20.6%) in primary care, continuing the upward trend observed in previous years. (Table 5a) Similarly, in secondary care there was an increase in use of trimethoprim (and sulphonamides) (4.9%) and nitrofurantoin (9.4%) compared to 2011, again continuing the upward trend since 2009 (Table 8). These increases in use of recommended agents are likely to reflect initiatives aimed at improving compliance with local prescribing guidelines. Reducing antibacterial use for prophylaxis of UTI (also referred to as medical prophylaxis ) remains a key clinical priority area for SAPG. The evidence base supporting antibacterial use for prophylaxis of UTI is not strong, as all studies were conducted before 2000, and no studies evaluated patients beyond one year. Recent evidence has shown there is a correlation between the number of prescriptions of antibacterials and resistance development, where multiple prescriptions increased the risk of a resistant E. coli 2-3 fold compared to a single course of treatment. 11 In 2012, 11 538 people received trimethoprim or nitrofurantoin on 6 occasions, which suggest extensive prophylactic use. This accounted for 24.0% of all cases where persons had received 6 courses of the same antibacterial in 2012. Over half (65.5%) of people receiving more than six courses of antibacterials were aged 65 years, and almost one in seven (14.8%) were recorded as having been resident in a care home at some point during 2012. This suggests that although prophylactic use of antibacterials for UTI may be common in care home residents, the practice may also be used in non-care home residents. It is recommended that initiatives to regularly review antibacterial prophylaxis of UTI in care home residents should be extended to include all patients in primary care prescribed multiple courses of antibacterials commonly used for UTI to ensure the ongoing clinical need is established. Antimicrobial resistance in urinary pathogens is increasing worldwide and in 2012 surveillance of antimicrobial resistance in urinary isolates was introduced in Scotland in order to detect emergence, cause and spread of resistance. Report on Antimicrobial Use and Resistance in Humans in 2012 7

Susceptibility patterns of urinary isolates are likely to reflect resistance to commonly prescribed antibacterials in the community and hospitals, and may provide an early indication of emerging resistance problems. At the time of producing this report, data from the UTI AMR surveillance were available from January 2012 to September 2013. Table 1: Number of organisms included in the surveillance of AMR of urinary tract infections from January 2012 - September 2013 Organism 2012 Q1 2012 Q2 2012 Q3 2012 Q4 2013 Q1 2013 Q2 2013 Q3 Grand Total Escherichia coli 3 513 3 541 3 431 3 378 3 391 3 494 3 585 24 333 70.91 Klebsiella pneumoniae Enterococcus faecalis 325 321 339 305 305 311 292 2 198 6.41 196 202 187 194 213 214 207 1 413 4.12 Proteus mirabilis 183 209 170 183 218 215 201 1 379 4.02 Enterococcus species Staphylococcus aureus Pseudomonas aeruginosa 158 137 103 64 41 94 77 674 1.96 78 78 76 53 79 86 93 543 1.58 81 74 87 76 65 93 58 534 1.56 Klebsiella oxytoca 71 82 73 71 76 69 79 521 1.52 Streptococcus species group B 54 52 61 59 53 56 61 396 1.15 Other 296 310 355 347 340 296 380 2 324 6.77 Total 4 955 5 006 4 882 4 730 4 781 4 928 5 033 34 315 100 % A total of 34 315 reports (Table 1) on non-duplicate urine samples were received from 13 NHS boards in this period. E. coli accounted for the majority of the reports (70.9%) and the numbers reported have remained high throughout the surveillance period. Susceptibility patterns in the E. coli urinary isolates (Table 2) were very similar to those observed in the E. coli bacteraemia isolates (Table 10 and Figure 16). E. coli resistance to the first-line agent trimethoprim increased from 38.0% in 2012 to 40.3% in 2013 (p<0.05). In 2013, resistance to nitrofurantoin also increased to 4.7% (a statistically significant increase as compared to 3.9% in 2012) (p<0.05). 8 Report on Antimicrobial Use and Resistance in Humans in 2012

Antimicrobial resistance to the majority of antibiotics tested (Table 2) significantly increased (p<0.05). The only exceptions to this were resistance to ciprofloxacin and gentamicin which remained relatively unchanged. It should be noted that the increase in resistance to both coamoxiclav and cefuroxime is attributable to changes from CLSI 12 to EUCAST 13 breakpoints. There are no intermediate categories for these two antibacterials in the EUCAST breakpoints; thus, all isolates previously categorised as intermediate are now categorised as resistant. Resistance to the third-generation cephalosporins varied from 4.8% for ceftazidime to 9.7% for cefotaxime in the urinary isolates. An important caveat to the AMR data on urinary isolates is the tendency to collect urinary samples only from patients who did not respond to initial empirical therapy. This creates a bias towards collecting the more resistant isolates which inflates the resistance proportions. This bias may be particularly strong in the samples from community patients. However, the resistance patterns in the isolates not responding to initial empirical treatment is of high relevance to prescribers for making decisions about treatment. Table 2: Resistance in E. coli urinary isolates from January 2012 - September 2012 and January 2013 - September 2013 Antibiotics Jan - Sep 2012 Jan - Sep 2013 p- value for difference (z-test) amoxicillin 54.8 57.7 0.038 ampicillin 57.8 59.8 0.008 cefotaxime 7.6 9.7 0.000 ceftazidime 3.8 4.8 0.000 cefuroxime 11.1 16.3 0.000 cephalexin 10.9 12.0 0.016 ciprofloxacin 16.6 17.2 0.242 co-amoxiclav 11.7 27.6 0.000 ertapenem 0.1 0.2 0.010 gentamicin 7.2 6.7 0.201 meropenem 0 0 0.045 nitrofurantoin 3.9 4.7 0.004 tetracycline 30.9 32.8 0.004 trimethoprim 38.0 40.3 0.000 There are limited international datasets for comparison, however, Sweden has been reporting on UTI surveillance since 1996. AMR comparisons between Sweden and Scotland should be interpreted with caution as sampling strategies differ in that Scotland monitors a substantially higher number of isolates than Sweden, while using a cross-sectional sampling approach. 14 In 2012, Sweden reported E. coli resistance proportions of 19.0% and 1.1% for trimethoprim and nitrofurantoin respectively. These proportions are lower than those found in Scotland however Sweden has reported an increasing resistance trend for both of these antibiotics since 1996. Report on Antimicrobial Use and Resistance in Humans in 2012 9

We will continue to develop the surveillance of antibacterial use and resistance in UTI. Along with AMR data from urinary samples the epidemiology of prescribing may inform improved patient care in this commonly encountered infection. In the previous report we stated that SAPG had developed a decision aid (algorithm) to support nursing, social care staff and prescribers with diagnosis and management of suspected UTI in older people. The algorithm was made available for use in primary care setting in May 2013. (http:// www.scottishmedicines.org.uk/files/sapg/sapg_decision_aid_for_uti_in_older_people_-_ May_2013.pdf). As part of the infection prevention and control approach to UTI guidance on preventing catheter associated UTI in acute settings was recently published by HPS. 15 Care improvement bundles have been shown to improve patient outcomes when collectively, reliably and continuously, can improve patient outcomes. This guidance contains an improvement bundle for preventing infection when inserting and maintaining a urinary catheter. The previous report recommended that analysis of the association between antibacterial use and resistance in urine isolates be undertaken to review the current recommendations on empirical treatment of UTI. A project has been funded by Chief Scientist Office (CSO) to develop the use of data linkage of existing datasets to identify a cohort of patients prescribed antibacterials commonly used for UTI, to describe the epidemiology of primary care prescribing for UTI and identify subsequent predictors of antimicrobial resistance and clinical outcome (hospitalisation and death). This work is intended to enable SAPG to review the recommendations for treatment of UTI. Epidemiology of Clostridium difficile Infection (CDI) and impact of antimicrobial stewardship Overall Scottish incidence rates of CDI have continuously shown a steep downward trend in patients aged 65 years in the period from 2008 to 2010 (Figure 4). From 2011 to 2012 the trend flattened, with the overall incidence rate for Scotland unchanged at 38 cases per 100 000 total occupied bed days. In the age group 15-64 years, the overall incidence rate for Scotland decreased 11% from 41 to 37 per 100 000 total occupied bed days between 2011 and 2012 (Figure 4). Substantial changes in the distribution of C. difficile ribotypes have been observed since 2008. There has been a remarkable decrease in the proportion of ribotypes 106, 001 and 027 (previously the most common ribotypes in Scotland), while the prevalence of the other major types: 078, 002, 005, 014, 015, 020 and 023 has increased. To date, all C. difficile isolates have been reported as susceptible to metronidazole and vancomycin. The majority of ribotypes isolated in 2012 from severe cases and/or outbreaks were resistant to clindamycin and cefotaxime. Cefotaxime resistance has increased in all the major ribotypes found in Scotland with overall resistance increasing from 62% in 2011 to 90% in 2012. Resistance to erythromycin and the fluoroquinolones (levofloxacin and moxifloxacin) decreased between 2011 and 2012. The changing patterns of antimicrobial susceptibility among C. difficile may influence the impact and efficiency of current antimicrobial use interventions. 10 Report on Antimicrobial Use and Resistance in Humans in 2012

It is not known why resistance to cephalosporins has increased substantially in all the major C. difficile ribotypes while the use of these agents have either declined in primary care or have not changed in secondary care between 2011 and 2012 (see Table 5a, Table 5b and Table 8). This will need to be investigated further; however, it remains evident that the use of any antimicrobial agent continues to be a risk factor for the development of CDI due to disturbance of the natural gut flora. Figure 4: Overall quarterly CDI incidence rates for Scotland (per 100 000 occupied bed days) in patients 65 years and 15-64 years for the period October 2006 to December 2012* Incidence rates per 100 000 total bed days 200 180 160 140 120 100 80 60 40 20 0 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 2006 2007 2008 2009 2010 2011 2012 Quarter and Year Age 65 and above Age 15-64 *surveillance in patients aged 15-64 years started April 2009 In 2012, reductions were observed in the use of cephalosporins (17.6%); penicillin combinations (mainly co-amoxiclav) (8.3%) and fluoroquinolones (8.0%) (Table 5a and Table 5b) in primary care, and continued the decreasing trend observed in previous years. In 2012, 9.2% of all antibacterial items dispensed in primary care were broad-spectrum antibacterials associated with a high risk of CDI. In secondary care (where 23.0% of the total antibacterial use is broad-spectrum agents) a different pattern of use in these broad-spectrum agents was observed in 2012 with an annual increase of 7.2%. The increase was largely accounted for by a 12.2% increase in co-amoxiclav and clindamycin (7.9%). In contrast, the use of cephalosporins and fluoroquinolones were unchanged from 2011 to 2012 (Table 8). As a result of the flattened trend in CDI incidence rates observed since 2011, there is a need to consolidate the previous reductions, and seek to drive further reductions in CDI rates. Report on Antimicrobial Use and Resistance in Humans in 2012 11

Good antimicrobial stewardship is one of the key factors that will help maintain the current low incidence rates, and this reinforces the importance of a review by SAPG and AMTs of the current local guidance for empirical use of broad-spectrum antibacterials in infection management. Although there have been reductions in the use of high-risk antibacterials in primary care between 2011 and 2012, this has not been observed in secondary care. It is interesting to observe a temporal association between the lack of reductions in the use of cephalosporins and fluoroquinolones in secondary care and the levelling of CDI incidence rates between 2011 and 2012. The extent of the impact of the increased use of co-amoxiclav (and to a lesser extent, clindamycin) on recent trends observed in CDI is also of concern. This will require further investigation as, depending on the strength of association, there may be the potential to reduce the burden of CDI even more as a result of further reductions in the use of high-risk antibacterials in secondary care. Having knowledge of the burden of CDI in the community may help to support understanding of the links between prescribing within primary and secondary care and CDI incidence rates. To help support this, HPS in collaboration with a subset of NHS boards is carrying out a pilot community CDI surveillance study; and results will be presented in the HAI Annual Report in 2014. In 2013, HPS commenced work to update the Guidance on Prevention and Control of Clostridium difficile Infection (CDI) in Care Settings in Scotland. 16 The guidance provides updated advice on management of CDI key aspects of prevention and control of CDI and an updated section underlining the importance of continued good antimicrobial stewardship. Prevention of Streptococcus pneumoniae by vaccination and impact on resistance patterns Streptococcus pneumoniae is a bacterium of major public health concern worldwide. It is a normal commensal of the upper respiratory tract; however it can cause a diverse spectrum of illness from upper respiratory tract infection to severe invasive disease including bacteraemic pneumonia, meningitis and septic shock. Immunocompromised individuals, including elderly, young children and people with serious health conditions, are at particular risk of developing pneumococcal infection. Worldwide, it is a leading cause of vaccine-preventable disease. Over 90 serotypes have been identified and their distribution within the human population varies greatly. The polysaccharide capsule differences between serotypes have been associated with differences in carriage, pathogenicity and case fatality. 17 Serotypes most commonly involved in pneumococcal disease or colonisation in infants are most frequently associated with antimicrobial resistance (namely beta-lactams, macrolides and fluoroquinolones). 18 This association with resistance may change over time with the increased use of the pneumococcal conjugate vaccine. Conjugate vaccines that are effective against the capsule have been available for the last 20 years. Polysaccharide antigens elicit a poor immune response in children less than two years of age, but conjugation to a protein overcomes this issue and invokes an effective immune response. It has been well documented that by immunising infants with the conjugate vaccine, the incidence of invasive pneumococcal disease is reduced across all age groups (herd immunity). Moreover, the pneumococcal polysaccharide vaccine (PPV23) is widely used in adults over 65 years of age and those with weakened immune systems. 19 12 Report on Antimicrobial Use and Resistance in Humans in 2012

The pneumococcal conjugate vaccines PCV 7 and PCV 13 were introduced in the Scottish childhood vaccination programme in September 2006 and March 2010, respectively. The PCV 13 vaccine is effective against serotypes 1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, 23F. There is the potential for resistant serotypes, not covered by the vaccines, to emerge and spread. In Scotland, the target of the national immunisation programme is for 95% of children to complete courses of childhood immunisations by 24 months of age. 20 Annual uptake rates by 24 months of age for primary courses of immunisation (including PCV 13) remain stable at 96% to 98%. Uptake rates have exceeded the 95% target for the last decade. In 2012, the most common Scottish serotypes causing bacteraemia were 7F, 8, 22F, 19A, 3, 15A, 33F, 6C, 11A, 23B, 16F and 35B accounting for 90.5% (371/410) of the typed isolates reported (Figure 5). Serotypes 6C, 8, 11A, 15A, 22F, 23B, 33F, 35B and 38 are not covered by the PCV. The PPV23 vaccine (used in adults) does not cover 6C, 15A, 23B or 35B. An increasing trend in the occurrence of serotypes 6C, 15A, and 35B has been observed since 2009 however, resistance against penicillin and erythromycin has remained stable. The trend, over the same time period, in serotypes 19A, 1 and 12F is decreasing. Figure 5: Distribution of the most common Scottish S. pneumoniae serotypes (%), 2010, 2011 and 2012 % Serotype 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 7F 8 22F 19A 3 15A 33F 6C 11A 23B 16F 35B 9N 23A 6A 12F 20 10A 38 1 Serotype 2010 2011 2012 Whilst resistance in the circulating Scottish pneumococcal serotypes has remained low and relatively stable since surveillance began in 2008, it should be noted that in Europe, 21 serotypes with combined resistance to beta-lactams and macrolides have been reported, particularly in children. We will continue to monitor the use of antibacterials in children and resistance in pneumococcal serotypes and report any emerging concerns to SAPG and key stakeholders. Report on Antimicrobial Use and Resistance in Humans in 2012 13