S44 Optimisation of control strategies and treatment

S10 ii. On a given continent, PVL-positive CA-MRSA have spread from country to country. For instance, in Europe, PVL-positive CA-MRSA of ST80 were recently detected in Slovenia, Romania and Croatia. iii. New PVL-positive CA-MRSA clones are emerging on different genetic backgrounds. While most of the clones described in 2003 had an agr3 background, the newly described clones are agr1 or agr2. Clone ST22 (agr1) has been found in Europe only and clone ST377 (agr1 with a type V SCCmec) was reported simultaneously in Europe and Australia. iv. PVL-positive CA-MRSA, which were initially susceptible to most antistaphylococcal antibiotics, have acquired new antibiotic resistance determinants, to gentamicin and ofloxacin for instance. v. The prevalence of PVL-positive CA-MRSA varies considerably from one continent to another (high for instance in the USA (~50%), Greece and Algeria, low in most European countries). vi. PVL-positive CA-MRSA are gradually causing an increasing number of hospital-acquired infections in countries where their prevalence is high. Conclusion: To counter this emerging global threat to public health, systematic surveillance of both hospital and community isolates is required, together with measures designed to limit their spread. S42 The impact of community-acquired MRSA in low-prevalence countries R. Skov (Copenhagen, DK) In the Nordic countries and the Netherlands, nosocomial MRSA has successfully been contained for more than 25 years keeping the prevalence of MRSA at <1% in these countries. This has been accomplished by enforcing preventive contact isolation of all hospitalised patients suspected or confirmed as MRSA positive. This has been facilitated by the fact that in these countries the predominant risk factor for patients for MRSA acquisition was hospitalisation abroad, making these at-risk patients relatively easy to identify. Additionally, secondary cases seldom arose after hospital discharge. The recent change in epidemiology where transmission of MRSA increasingly occurs outside hospitals has changed the situation, and an increasing number of MRSA has been reported in all the above mentioned countries. Since the mid-1990s the number of new cases of MRSA in Denmark has increased by 10-fold. The majority of these MRSA has been identified in primary healthcare (community onset) and, in approximately 33% of the cases, the patient had had no connection with healthcare institutions for at least one year prior to the diagnosis. These cases have therefore been considered as community acquired (CA-MRSA). CA-MRSA infections have especially been reported from children and younger adults. Skin and soft tissue infections dominate and represent approximately 90% of the cases. Approximately 30% of CA-MRSA cases have been reported in persons with another ethnicity than Danish. This could indicate that there is a continued influx of MRSA from high prevalence areas into Denmark. Strains belonging to the ST80IV complex have been the most frequent cause of CA-MRSA in Denmark, however, an increasing diversification has recently been observed including almost all strains known as associated with CAMRSA. During the same time period the number of MRSA diagnosed in hospitals also increased. This has happened despite continued use of contact isolation of MRSA positive patients during. Hospitalisation of MRSA carriers not identified on admission due to lack of risk factors is probably one important factor responsible for the increased number of nosocomial MRSA in Denmark. An increase in the prevalence of MRSA outside hospitals including in otherwise healthy persons, is thus of great concern as this not only gives rise to CA-MRSA infections but also inevitably will increase the risk of nosocomial MRSA infec-tions and outbreaks. In order to keep Denmark as a low prevalence country we believe that MRSA should be contained and eradicated not only in hospitalised patients but also when diagnosed in the community. This has lead to new

17th ECCMID / 25th ICC, Oral presentations national guidelines for preven-tion and eradication of MRSA including making both carriage and infection with MRSA mandatory reportable. S43 How, when and whom to screen K. Christiansen (Perth, AU) The first question to ask when considering any screening programme for the identification of CA-MRSA carriage is why? The importance of epidemiological data cannot be underestimated. Data on prevalence can inform empirical therapy requirements or help in the design of intervention strategies. Prevalence can be established for the population in general, or, specific subgroups such as children, military personnel, indigenous or disadvantaged populations can be targeted. Community intervention measures similar to the ‘search and destroy’ techniques used in hospitals would require large-scale screening. A narrower approach concentrating on high-risk groups to enable programmes such as education on hygiene to reduce transmission would allow a more targeted screening programme. A second more pragmatic area for screening is on admission or during hospitalisation. In some areas in the world there is a blurring of the boundaries of CA-MRSA and HA-MRSA. These organisms have now been reported as a cause of hospital acquired bacteraemia, surgical and prosthetic joint infection and to colonise neonates and adult patients in hospitals. Targeted screening of high-risk patients or all patients on entry into high-risk units are strategies to prevent CA-MRSA from becoming endemic in hospitals. Screening to control recognized outbreaks entails identification of the reservoir to enable appropriate isolation, cohorting and decolonisation measures. The value of screening healthcare workers for carriage of CA-MRSA has yet to be established. How to screen is the next question. Which site(s) give the best yield? Which laboratory method should be used? The ideal test would have high negative and positive predictive values, be rapid, inexpensive, applicable to the routine laboratory, and have high through put capabilities. Currently used tests are traditional microbiological culture and identification methods, use of selective media including various types of CHROMagar and molecular techniques using real time PCR. The requirement for broth amplification increases the sensitivity for some of these methods but also increases the time to identification. There are many different CA-MRSA clones in the world, in Western Australia alone there are over 40 different multi locus sequence (MLST)/SCCmec CA-MRSA clones. Any molecular test must therefore have the ability to identify all clones in diverse regions of the world. The last questions are when and whom to screen. The answer is very much dependent on the purpose of the screening. The range can be from large-scale periodic population screening for epidemiological purposes to continual hospital unit, high-risk patient screening. There are no established guidelines or recommendations for screening for CA-MRSA. Many questions have yet to be answered particularly, whether control of transmission in the community is possible. By looking at what evidence is available we can move a step closer to control of this very significant community pathogen. S44 Optimisation of control strategies and treatment S. Cosgrove (Baltimore, US) Rates of infections caused by CA-MRSA continue to increase, causing over 50% of skin and soft tissue infections seen in the emergency department in a recent report. Strategies for control of CA-MRSA in patients and their families include use of infection control techniques in the household as well as decolonisation with mupirocin, chlorhexidine, and antibiotics in some cases. Given that CA-MRSA has been now been implicated in healthcare-associated infections, efforts to reduce the spread of this organism within the healthcare environment are also important and include appropriate hand-hygiene, isolation, and environmental decontamination. Incision and debridement remains critical in the management of most CA-MRSA infections. The use of antibiotics is indicated in many CA-MRSA infections and therapeutic

Pathogenesis and epidemiology of invasive pneumococcal infections options include older agents, such as clindamycin, trimethoprimsulfamethoxazole, tetracyclines, and rifampin, and newer agents such as linezolid. In addition, new agents are in development.

Pathogenesis and epidemiology of invasive pneumococcal infections S45 Recent discoveries in the pathogenesis of pneumococcal pneumonia D.H. Dockrell (Sheffield, UK) Pneumococcal pneumonia is a pathogen of global significance and represents a leading cause of infection-related mortality at all ages. Recent developments in the field of pneumococcal pathogenesis reflect significant developments in both genomics and proteomics. The breadth of microbial genetic diversity is increasingly being recognized. Our understanding of the importance of traditional virulence factors such as polysaccharide capsule, pneumolysin and generation of hydrogen peroxide is evolving while important roles for factors such as pneumococcal pili are emerging. Genetic screens are identifying increasing roles for previously poorly characterised proteins involved in metabolism and transport of key molecules. The importance of the physical state of the bacteria in different tissue compartments is also becoming evident. An improved range of models of various aspects of infection ranging from colonisation and sub-clinical infection to fulminant pneumonia and invasive disease are allowing assessment of both microbial and host factors central to disease pathogenesis. The range of receptors involved in the innate response to pneumococci in the respiratory tract is starting to be appreciated and includes Tolllike receptors and receptors primarily involved in the phagocytosis of bacteria. The interaction between soluble factors and the resident phagocytes of the lung is being clarified and the complexity of cytokine networks involved in the innate host response is being elucidated. Important roles for T-lymphocytes, Natural Killer cells and dendritic cells in the response to pneumococci are being identified. Furthermore our understanding of the regulation of the inflammatory response and of the significant role of apoptosis in the regulation of the inflammatory response has lead to important insights into how cell survival and outcome of infection are closely linked. These observations are enabling a more scientific interpretation of many of the central clinical features of pneumococcal pneumonia and may encourage novel approaches to therapy to improve disease responses. S46 Invasive disease potential of pneumococcal clones carried by children H. de Lencastre, R. S´a-Le˜ao, M. Ramirez, J. Melo-Cristino (Oeiras, Lisbon, PT) Streptococcus pneumoniae (the pneumococcus) continues to be a leading cause of morbidity and mortality worldwide particularly among young children, the elderly and the immunocompromised of all ages. Infection is preceded by colonisation of the nasopharynx, which is the ecological niche of pneumococci. In most individuals colonisation is asymptomatic and does not evolve to disease. The carrier state is more frequent in young children and may reach over 70% in some populations such as those attending day-care centres. Indeed, this latter group has been found to be a major reservoir of pneumococci playing a key role on the amplification and transmission of the bacteria to other individuals. Of interest, several drug-resistant clones internationally disseminated have been found to be frequently carried by this population. In recent years, there has been some debate on the relative contribution of serotype and genetic background to the invasive disease potential of pneumococcal strains. We will present a study that addresses this question. The study was conducted in Portugal and results from the detailed comparison of two large collections of pneumococcal isolates: a group of over 450 invasive disease isolates and a group of over 750 colonisation isolates

S11 collected during the same time period which were characterised in detail by antibiogram, serotype, PFGE macro-restriction profiles, and multilocus sequence typing. The invasive disease potential of serotypes and clones will be discussed. S47 Epidemiology of invasive pneumococcal infections A. Brueggemann (Oxford, UK) Pneumococci cause a spectrum of diseases in humans, from reasonably mild diseases like sinusitis and conjunctivitis to potentially lifethreatening diseases like meningitis and bacteraemia. The current conjugate vaccines are ideally aimed at protecting against all pneumococcal disease, but have been very successful at preventing invasive pneumococcal disease. Understanding invasive disease epidemiology, both preceding vaccine implementation and after vaccine introduction, is crucial to the design and development of future vaccines. Several recent studies have shown that pneumococcal serotypes differ in their invasive disease potential, and this has particular relevance for the selection of serotypes to include in future conjugate vaccines. It is also essential to understand the serotype-specific changes that have occurred subsequent to conjugate vaccine implementation in the USA, as a model for what might occur post-vaccine implementation in other countries. S48 Recent discoveries in prevention of pneumococcal disease K. Klugman (Atlanta, US) The recent history of pneumococcal disease prevention is dominated by the development in the 1990’s of conjugate pneumococcal vaccine for administration to infants, and its implementation starting in 2000 in the USA. The vaccine as currently formulated affords protection against invasive pneumococcal disease (IPD) due to the 7 most common serotypes causing IPD among children less than 2 years of age prior to vaccine introduction in the USA. This formulation is not optimal for many developing countries where serotypes such as 1 and 5 are important. Global, second generation vaccines therefore include at least 10 serotypes. The importance of prevention of pneumococcal disease extends beyond IPD to pneumonia, meningitis, otitis media, prevention of antibiotic-resistant infections, and even prevention of mortality in young infants. Each of these presents specific challenges. The vaccine has been shown to prevent between 25% and 37% of all cause X-ray confirmed pneumonias and this implies a level of protection against pneumococcal pneumonia due to vaccine serotypes in excess of 70%. This reduction in pneumonia has translated into a reduction in allcause mortality due to the vaccine of 16% in rural Africa. The major preventable burden of disease in Africa may be among HIV infected children where the vaccine has also been shown to reduce pneumonia burden. An innovative aspect of the vaccine is the fact that it may reduce the pneumococcal super infections that follow viral respiratory infections, such as influenza, and conjugate vaccination of children may thus be a useful adjunct to pandemic influenza planning preparedness. The interruption of carriage of vaccine serotypes has been shown to reduce IPD in adults in the USA due to herd immunity so the population benefit in an influenza pandemic may extend beyond protection just of immunised children. Herd immunity has also recently been shown to protect young un-immunised infants from IPD. Conjugate vaccine has had little impact on all cause otitis media in randomised trials, but post-marketing studies in the USA suggest that healthcare utilisation for otitis episodes has reduced significantly post vaccine introduction suggesting a major role for herd immunity and allowing physicians to feel more comfortable in a “wait and see” attitude to otitis once the prospects of significant complications due to the pneumococcus are likely to be rare. The conjugation of pneumococcal polysaccharides to Haemophilus influenzae protein D has allowed the development of a more efficacious vaccine against otitis media. There has been a dramatic reduction in antibiotic resistance among vaccine serotypes causing IPD in the USA, in both children and adults, but resistance, driven by continuing high levels of antibiotic use, is now selecting resistance in non-vaccine