Epidemiology of MRSA in Scotland

Abstract

Staphylococcus aureus (S. aureus) is a bacterium that commonly colonises the skin and nares of around one third of otherwise healthy individuals. While colonisation is benign, S. aureus can cross skin and mucosal barriers to cause infections that manifest as clinical disease. Clinical outcomes are diverse and range from mild, non-complicated and often self-limiting skin and soft tissue infections (including boils, abscesses and cellulitis) to more severe and life-threatening conditions including pneumonia, toxic shock syndrome and bacteraemia. Medication isn’t always needed for mild S. aureus infections as often they resolve with time but, for severe or persistent cases, antimicrobial treatment is generally required. Following decades of widespread and intensive usage of topical, enteral and parenteral antimicrobials to treat S. aureus infections; AMR has become an established and ubiquitous problem in the treatment of infections caused by this microorganism, especially when in the methicillin resistant form (i.e. MRSA). The aim of this thesis was to examine aspects of S. aureus epidemiology (including MRSA and methicillin-sensitive S. aureus (MSSA)) in Scotland using statistical methods and data from several large public health databases. More specifically this involved: descriptions of spatial and temporal trends of morbidity and mortality; comparisons of epidemiological and molecular attributes (including antimicrobial resistance) of (1) MSSA and MRSA, and (2) the dominant clones of MRSA (i.e. EMRSA-15 and EMRSA-16); descriptions of spatial and temporal trends of antimicrobial prescribing in primary and secondary care and any associations between prescribing rates and MRSA antimicrobial resistance; and carrying out a hospital-level risk factor analysis of MRSA, testing hypotheses that hospital size, hospital connectivity (through shared transfer patients) and hospital category have an effect on hospital-level incidences of MRSA in mainland Scottish hospitals. Results showed that total S .aureus bacteraemia and MRSA bacteraemia in Scotland statistically declined over time (p<0.0001), but MSSA bacteraemia did not (p>0.05). While combined mortality rates (i.e. all MSSA deaths (both primary and secondary cause), or all MRSA deaths (both primary and secondary cause)) mirror these findings; case-fatality ratios (CFR) show no declines over time for either MRSA or MSSA. Results also show that several epidemiological factors point towards a predominant community source for MSSA isolates (i.e. outside healthcare) and hospital source for MRSA. Evidence for this included: (1) the lack of resistance genes in the MSSA population, (2) MRSA was more associated with long-term care and high-risk patients in the specialties care of the elderly, high dependency units /intensive care units (HDU/ICU), and surgery and conversely MSSA with specialties that commonly served outpatients, and (3) the abundance of non-EMRSA-15/non- EMRSA-16 ‘other’ clones in the MSSA population as compared with the hospital-associated CC22 (EMRSA-15) and CC30 (EMRSA-16) clones. EMRSA-15 was by far the most dominant MRSA clone in Scotland with EMRSA-16 declining significantly and non-EMRSA-15/non-EMRSA-16 clones causing an increasing number of infections (over the time period 2003-2013). EMRSA-16 was resistant to a larger number of antimicrobials than EMRSA-15, typically 9 versus 5, and while resistance varied for EMRSA-16 over the study period, resistance remained stable for EMRSA-15. There was little difference between clinical and screening MRSA isolates. Analyses of antimicrobial prescribing showed that prescribing rates of several drugs increased over time (2003-2013). Prescribing was far higher in primary care settings than in secondary care, although this differed between antimicrobials. Significant positive associations between prescribing and resistance rates were found for gentamicin (pr - p<0.0001, se - p<0.0001) and trimethoprim (pr - p<0.01, se - p<0.0001) in both primary (pr) and secondary (se) care, and clindamycin (p<0.0001) in primary care only. Finally, in Scotland there is a threshold of connectivity above which the majority of hospitals, regardless of size, are positive for MRSA. Higher levels of MRSA are associated with the large, highly connected teaching hospitals with high ratios of patients to domestic staff. While there were a number of data limitations, this body of work provides a better understanding of the epidemiology of S. aureus including MRSA in Scotland.