Since 2001 the Swedish Institute for Infectious Disease Control (SMI) and the Swedish Strategic Programme Against Antibiotic Resistance (Strama) have presented annual reports on antibiotic use and resistance in human medicine that are co-published with a corresponding veterinary report.
Data on antibiotic consumption are collected from the National Pharmacy´s database and the Swedish register of prescribed drugs. Resistance data are from notifications according to the communicable disease act and several national surveillance networks.
A small decrease in antibiotic sales was observed in 2008. Most notably, the consumption of quinolones and cephalosporins decreased. This is in all probability the result of new guidelines (a) for the treatment of lower urinary tract infections in women, (b) for the treatment of community-acquired pneumonia and (c) to reduce the selective pressure on ESBL-producing bacteria in the hospitals.
There were three major findings of epidemiological and microbiological interest: (a) an outbreak of vancomycin-resistant enterococci detected during 2007 in Stockholm continued and resulted in outbreaks in several other counties, (b) cases of ESBL-producing E. coli and K. pneumoniae continued to increase and are now more frequent than MRSA and (c) the transmission of MRSA in the healthcare sector is low and seems to have stabilised. This could be a result of extensive efforts to promote compliance with basic hygiene principles including alcohol rub hand hygiene.
Our data indicate that we continue to successfully combat major outbreaks of resistant bacteria. However, ongoing outbreaks of vancomycin-resistant Enterococcus faecalis and Enterococcus faecium (VRE) in hospitals and a general increase in ESBL-producing Enterobacteriaceae are of concern.
Convincing evidence points at the relation between increased antibiotic consumption and higher levels of antibiotic resistance . To date, Sweden is among the countries with the lowest rates of antibiotic resistance as reported to the European Antibiotic Resistance Surveillance System (EARSS)  and also among the countries with a low antibiotic consumption as reported to the European Surveillance of Antimicrobial Consumption (ESAC) .
The Swedish Pharmaceutic monopoly has formed the basis for the collection of high quality antibiotic consumption data. Similarly, a long tradition and interest among clinical microbiologists have made it possible to collect highly standardised results from antibiotic susceptibility testing (AST) of clinically relevant bacterial isolates. Since 2001 these data have been reported in annual reports on antibiotic use and resistance in human medicine, the SWEDRES reports. These reports, accessible via the institute’s website, have been presented by the Swedish Strategic Programme Against Antibiotic Resistance (STRAMA) together with the Swedish Institute for Infectious Disease Control (SMI) and have been co-published with the corresponding veterinarian report SVARM (Swedish Veterinary Antimicrobial Resistance Monitoring) prepared by the National Veterinary Institute. In this paper we summarize the major findings from SWEDRES (Swedish Antimicrobial Utilisation and Resistance in Human Medicine) 2008 .
Surveillance of antibiotic consumption
The data systems of the National Corporation of Swedish Pharmacies (Apoteket AB) and the Swedish prescribed drug register (operated by the National Board of Health and Welfare since 2005), are the main sources for data on antibiotic consumption. The Anatomical Therapeutic Chemical (ATC) classification system recommended by the World Health Organization (WHO) has been used since 1988. The sales of drugs are presented as number of defined daily doses (DDDs).
Out-patient care data are based on the sales of drugs dispensed on prescription (since 1996) or by ApoDos (individually packed doses of drugs often dispensed to elderly, since 1999) by all Swedish pharmacies. The individually-based Swedish prescribed drug register supplies data on the number of individuals treated with at least one course of antibiotics during a specific period of time and on the number of purchases per person.
Hospital data is based on drugs delivered by the hospital pharmacies to the hospital departments. The sales are expressed as cash value, number of packages and number of DDDs and the denominators (number of admissions or number of bed-days) are obtained from the Swedish Association of Local Authorities and Regions.
Antibiotic susceptibility testing
Phenotypic methods (disk diffusion or Minimum Inhibitory Concentration (MIC)) are used as the routine methods for susceptibility testing. They are performed and interpreted according to criteria for sensitive-intermediate-resistant (SIR)-categorization provided and regularly updated by the the Swedish Reference group for Antibiotics (SRGA). The phenotypic routine methods sometimes need to be complemented by methods for gene detection, e.g. in the case of methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus faecalis and Enterococcus faecium (VRE), or by MIC determination, e.g. in cases of reduced susceptibility to beta-lactam antibiotics in pneumococci and Haemophilus influenzae. Direct methods for detection of beta-lactamase production are recommended for Haemophilus influenzae and Neisseria gonorrhoeae.
Each laboratory is responsible for its own internal and external quality assurance and quality control programme. Recommendations for internal quality control are provided by SRGA. External quality control is often done by participation in United Kingdom National External Quality Assessment Service (UK-NEQAS) and/or other international programs, whereas quality assurance is one of the features of the Swedish ”100-strains” or RSQC (Swedish Combined Resistance Surveillance and Quality Control) programme.
Surveillance of antibacterial resistance
There are five branches for surveillance of antibiotic resistance in Sweden. The first is based on notifications according to the Communicable Disease Act (CDA). Penicillin nonsusceptible Streptococcus pneumoniae (PNSP) with penicillin G MIC > 0.5 mg/L became notifiable according to the CDA in 1996. In 2000 mandatory notification of MRSA and VRE was introduced, which was followed by extended-spectrum beta-lactamase- (ESBL) producing Enterobacteriaceae in 2007. Both symptomatic cases and asymptomatic carriers identified during contact tracing or screening are notified with full identity by both the laboratory and the clinician (except for ESBL-producing Enterobacteriaceae which are only notified by the laboratories). As an important complement to the notifications, most newly detected strains of MRSA, VRE and PNSP are sent to SMI for epidemiological typing.
Secondly, in 1994 a model for the concomitant surveillance of antimicrobial resistance and quality assurance of antimicrobial susceptibility testing was devised (RSQC surveys). Each year all laboratories are asked to collect quantitative data (zone diameters) for a predefined set of antibiotics on 100-200 consecutive clinical isolates of a number of bacterial species. Data are submitted and presented using the web-based software ResNet (http://www.srga.org/ResNet_sok.htm). Once submitted and accepted, the results from each laboratory/county are displayed as a resistance frequency on a map. A graph showing the distribution of zone diameters or MICs together with the relevant demographic data is found behind each frequency figure. The ResNet software is also available for other sets of aggregated, quantitative data of relevant collections of isolates, e.g. invasive isolates from several laboratories that form the Swedish part of the EARSS.
The Swedish participation in EARSS forms the third branch of the surveillance programme. Twenty-one laboratories, covering approximately 75 % of the population, participate continuously and deliver data on invasive isolates of seven bacterial species: Staphylococcus aureus, Streptococcus pneumoniae, Escherichia coli, Enterococcus faecalis, Enterococcus faecium, Klebsiella pneumoniae and Pseudomonas aeruginosa.
The fourth branch comprises a subset of 11 laboratories covering a population of 3.7 millions (> 40% of the population) that supply data on invasive isolates from all positive blood cultures.
Finally, reference laboratories and sentinel laboratories deliver data from susceptibility testing of gastrointestinal pathogens such as Salmonella, Shigella, Campylobacter jejuni/coli and Helicobacter pylori and also on Neisseria gonorrhoeae and N. meningitidis.
After three consecutive years with increasing total antibiotic sales, a 1.6% decrease was seen in 2008. (Fig. 1). Tetracyclines and different kinds of penicillins are the most commonly used classes of antibiotics in primary health care. There is a great variance in terms of number of prescriptions and preference of substance between counties. One third of all children aged 0 to 6 years were treated with at least one course of antibiotics in 2008.
Fig. 1. Total antibiotic use in the Nordic countries
Figure 2 illustrates the geographical variation in over all prescription and, in particular, prescription of penicillins with an extended spectrum to children aged 0 to 6 years, which varies from 20 % in the counties with the lowest use and 40% in the high prescribing counties.
Fig. 2. Variation in prescriptions of penicillins to children aged 0-6 years (number of prescriptions/1000 children) in different Swedish counties, 2008
Pivmecillinam and nitrofurantoin, the recommended first-line antibiotics for uncomplicated lower urinary tract infections (UTI) in women aged 18-79 years, now comprise more than 60% of the prescriptions (Fig. 3).
Fig. 3. Antibiotics commonly prescribed to women aged 18-79 years for uncomplicated lower urinary tract infections in Sweden
The use of fluoroquinolones is still high, particularly among the elderly.
The total level of antibiotics sold only to hospitals has been fairly constant during the last three years when related to patient admissions and patient-days. The use of different groups of antibiotics in hospitals varies between the counties, with narrow-spectrum antibiotics such as penicillin V and G accounting for between half and one third of the use. Beta-lactamase-sensitive penicillins increased and cephalosporins decreased, and reached the same level by the end of 2008. The use of fluoroquinolones was also decreasing in hospital care.
Dentists accounted for approximately 7% of all antibiotic prescriptions in community care, 80 % of which was penicillin V. The over all prescription is increasing for all antibiotic classes used by dentists; penicillin V, amoxicillin and clindamycin.
Fig. 4. Notifications of antibiotic resistance according to the Swedish communicable disease act
A total of 1,307 cases of MRSA were notified in 2008 (Fig. 4) corresponding to a national incidence of 14.1 cases/100,000 inhabitants. This represented a 16% increase compared to 2007. Sixteen invasive isolates were reported corresponding to 0.7 % of all blood isolates of Staphylococcus aureus. More than half of the reported cases (665 cases) were domestic and one-third (450 cases) imported. The distribution of healthcare- and community-acquired MRSA among domestic cases is shown in Figure 5.
Fig. 5. Place of acquisition of MRSA among 4,033 domestic cases notified in Sweden, 2000-2008
Epidemiological typing of all MRSA isolates has been performed by spa-typing since 2006. The five most commonly encountered spa-types in 2008 were t002 (n=132), t008 (n=113), t044 (n=107), t019 (n=54) and t032 (n=51). The prevalence of MRSA with panton valentine leukocidin (PVL) toxin was slowly increasing and was present in all or a majority of isolates with the common spa-types t008, t044, and t019. Staphylococcus aureus from wound infections (RSQC programme) were susceptible to antibiotics in > 95% of the cases, the only exception being fusidic acid resistance which was decreasing but still above 5%.
In 2008 there were 565 notifications of PNSP (benzylpenicillin MICs of >0.5 mg/L) as illustrated in Figure 4. Most cases were from nasopharyngeal cultures. Only 19 (3.4%) were from invasive infections. The majority of PNSP cases were found in the age group 0–4 years. Rates of PNSP were lower among invasive isolates than in nasopharyngeal isolates from the RSQC programme. Resistance to macrolides was 5–6% in both types of isolates.
In 2008 there were 618 notified cases of VRE, almost 12 times more cases than in 2007 (Fig. 4). This high number was attributable to the spread of vanB-carrying Enterococcus faecium. One strain, which was new according to epidemiological typing by Pulsed Field Gel Electrophoresis (PFGE), was detected in Stockholm county already in 2007, but became epidemic in the counties of Halland and Västmanland during 2008. This strain gave rise to clinical infection in only 10% of the cases and was found as a colonizer of the faecal flora in the majority of cases. This new epidemic strain also appeared in blood cultures, giving a rate of vancomycin resistance of 1.5% as reported to EARSS.
Data on Streptococcus pyogenes and Streptococcus agalactiae were obtained from eleven laboratories delivering data on all blood culture isolates. Only 196 and 107 invasive isolates, respectively, were recorded. Macrolide resistance was found in 0.5% and 6.5%, respectively. Fifteen percent of S. pyogenes were resistant to tetracycline.
Haemophilus influenzae was re-entered into the RSQC programme in 2008 after three years without reporting. Resistance rates for penicillin and trimethoprim-sulfamethoxazole had increased dramatically and were > 20%. One third of the patients with beta-lactamase-producing isolates were children 0–9 years while the remaining isolates were evenly distributed among all other age groups. Haemophilus influenzae was rarely found among blood isolates, only 63 cases in 2008 according to data derived from the eleven laboratories. Sixteen of the isolates (25%) were beta-lactamase-producing.
A total of 2,957 cases of Enterobacteriaceae with ESBL, predominantly Escherichia coli and Klebsiella pneumoniae, were notified during 2008 (Fig. 4) giving an average national incidence of 32 cases per 100 000 inhabitants. There was a great geographical variation in incidence ranging from almost 60/100,000 in Uppsala county (where a large hospital outbreak of K. pneumoniae was under investigation), down to approximately 15/100,000 Norbotten. A 28% increase of ESBL cases was noted in 2008 as compared to 2007. Most ESBLs were found in urine samples (70%) and the most commonly reported species was Escherichia coli (84%). Isolates with ESBLs, most often of the CTX-M-type, were often multiresistant.
In E. coli, mainly derived from urinary tract infections, ampicillin resistance caused by the production of plasmid-mediated beta-lactamase (most often of TEM-type) was increasingly found in both blood isolates and urine isolates (32% and 29%, respectively) in 2008. The level of resistance to third generation cefalosporins increased to 2.2% among blood isolates, and in the majority of these cases the resistance was caused by CTX-M type plasmid-mediated ESBLs. This resistance was often accompanied by resistance to many other antibiotics, e.g. aminoglycosides and fluoroquinolones. Resistance to fluoroquinolones has increased every year and was almost the same in urine as in blood isolates (13 vs. 14%) in 2008.
Other gram-negative bacteria that have been monitored in the RSQC programme and in the EARSS network are Klebsiella pneumoniae and Pseudomonas aeruginosa. The levels of resistance for the antibiotics tested were comparable between the two surveillance programmes for each of the microorganisms. Approximately 2% of Klebsiella pneumoniae were cephalosporin resistant and ESBL-producing. In 2007 the first isolate of Klebsiella pneumoniae with KPC-2, a carbapenemase, was detected in Sweden. In 2008 at least one more isolate with a KPC beta-lactamase was identified, and one isolate with a metallo-beta-lactamase of the VIM-type. In all these cases a history of hospital care in the south of Europe was reported. In Pseudomonas aeruginosa, the prevalence of carbapenems resistance was approximately 5% and fluoroquinolone resistance 10%.
The decrease in antibiotic sales in Sweden 2008 is not fully understood. Antimicrobial resistance and healthcare issues related to the use of antibiotics and local outbreaks have received considerable attention in the media over the past few years and may have contributed to the decline.
The change in prescribing patterns for uncomplicated female UTIs reflect systematic information campaigns by local STRAMA groups and drug committees that have been ongoing for several years. In 2008 information campaigns were coordinated at the national level.
Analysis of antibiotic use among the elderly is complicated due to the fact that medicines, including antibiotics, are increasingly often dispensed (rather than prescribed) to residents in nursing homes. Sales of medicines to many elderly are therefore erroneously allocated to the statistics for hospital care.
While the overall hospital use of antibiotics has been relatively stable, a desired shift from broad-spectrum cephalosporines and fluoroquinolones to the penicillin group is under way. This is a remarkable shift in our long tradition of cephalosporin use. The driving force behind the change is the rapidly escalating problem with ESBL-producing Enterobacteriaceae and the subsequent launching of the “STRAMA programme against ESBL” in 2007 . Our point prevalence studies in 2003, 2004, 2006 and 2008, confirm the considerable decrease in the use of cephalosporins for the treatment of uncomplicated community-acquired pneumonia.
Sweden is still one of the few countries with less than 1% of MRSA among invasive Staphylococcus aureus (from EARSS reports). The number of domestically acquired healthcare-related MRSA cases is no longer increasing. This has also been observed in some other European countries with an initially much higher prevalence of invasive MRSA [2,6,7]. PNSP have decreased in annual incidence rates per 100,000 population from around 10 in 1997 to between 6 and 8 since 2000, and severe penicillin resistance is very rare. An outbreak of VRE recognized in Stockholm county during the autumn of 2007 unfortunately disseminated to several other counties and is now causing secondary outbreaks including the geographically distant Halland county. Despite major efforts, the spread is still not under control in all places.
When Haemophilus influenzae re-entered in the RSQC programme 2008, strikingly high penicillin and trimethoprim-sulfamethoxazole resistance rates (>20 % compared to approximately 10% during the previous years) were seen. There was no correlation to age. The explanation for this change is still unclear and under investigation.
The numbers of Enterobacteriaceae with ESBL continue to increase dramatically. Most isolates contain ESBLs of the CTX-M-type and are also often multiresistant, seriously limiting the therapeutic options. Several outbreaks and also deaths related to ESBL-producing bacteria occurred in 2008.
Despite major efforts by SMI and STRAMA, the success in changing some prescribing habits, the decrease in the use of antibiotics in some age groups and the availability of good quality data from microbiological laboratories and pharmacies, resistance rates are increasing and significant outbreaks with resistant intestinal bacteria including VRE and ESBL-producing Enterobacteriaceae are seen in hospitals and institutions. Surveillance of antimicrobial resistance needs to be intensified. Systems for early detection of extreme resistance and outbreaks of resistant bacteria must be developed. Infection control measures in hospitals and institutions to counteract the dissemination of bacteria in general and resistant bacteria in particular, must be further strengthened and compliance with recommendations for treatment needs to be improved. This was addressed in a recent report to the Swedish government.
- Sande-Bruinsma van de N, Grundmann H, Verloo D, Tiemersma E, Monen J, Goossens H, Ferech M. European Antimicrobial Resistance Surveillance System Group; European Surveillance of Antimicrobial Consumption Project Group. Antimicrobial drug use and resistance in Europe. Emerg Infect Dis 2008;14:1722-30.
- EARSS Annual Report 2007. EARSS Management Team, members of the Advisory Board, and national representatives of EARSS, Bilthoven, The Netherlands, October 2
- Muller A, Coenen S, Monnet L, Goossens H. ESAC project group. European Surveillance of Antimicrobial Consumption (ESAC): outpatient antibiotic use in Europe, 1998-2005. Euro Surveill. 2007;12:E071011.1.
- Struwe J, Olsson-Liljequist B (editors). SWEDRES|2008 – A Report on Swedish Antimicrobial Utilisation and Resistance in Human Medicine. Strama, The Swedish Strategic Programme against Antibiotic Resistance, and the Swedish Institute for Infectious Disease Control. Stockholm, 2007.The Swedish Strategic Programme against Antibiotic Resistances (Strama). ESBL in enteric bacteria. Proposed action plan – November 2007.
- MRSA bloodstream infections continue to fall in East of England. Health Protection Agency, London, UK. 19 March 2009.
- Campanile F, Bongiorno D, Borbone S, Stefani S. Hospital-associated methicillin-resistant Staphylococcus aureus (HA-MRSA) in Italy. Ann Clin Microbiol Antimicrob 2009; 8:22.