The clinical impact of methicillin-resistant Staphylococcus aureus on morbidity, mortality and burden of disease

Abstract

The aim of this thesis was to evaluate the clinical impact of methicillin-resistant Staphylococcus aureus [MRSA] infections on the total burden of disease. A guideline on empirical antimicrobial eradication of MRSA in carriers was developed based on a systematic review of literature. A distinction was made between uncomplicated and complicated carriage depending on presence or absence of skin lesions, foreign body material, mupirocin resistance or detection of MRSA exclusively outside the nose. First choice empirical treatment in uncomplicated carriage is a combination of mupirocin nasal ointment and disinfecting soap. First choice empirical treatment in complicated carriage is mupirocin nasal ointment and disinfecting soap together with a combination of two antibiotics. We evaluated the effectiveness of this guideline in a prospective cohort study in 18 Dutch centres. Eradication occurred more frequently among those treated according to the guideline. However, adding systemic antibiotic therapy to topical treatment in all patients with extranasal carriage, chronic pulmonary disease or ADL-dependence may further increase treatment success. We quantified changes in annual trends of nosocomial bacteraemias [NB] in a Dutch tertiary care centre. We concluded that the incidence density is sensitive to changes in hospital number of patient-days over time and most accurately reflects the risk per day in the hospital. We hypothesized that an increased incidence density of NB caused by MRSA was associated with a reduced incidence of NB caused by other pathogens, resulting in similar trends in total incidence rates of NB in hospitals with and without MRSA endemicity. We therefore calculated annual incidence densities of NB of MRSA and all other pathogens in 7 hospitals with, and 5 hospitals without emergence of MRSA from 1998 through 2007. Originating from a comparable total burden of NB in 1998, the incidence density of MRSA and non-MRSA antibiotic-resistant bacteria increased more in MRSA endemic hospitals, resulting in a 29% higher total burden of NB after 10 years, demonstrating that emergence of infections with antibiotic-resistant bacteria occurs in addition to infections caused by antibiotic-susceptible bacteria. We quantified inadequate antibiotic therapy for S. aureus bacteraemia in Western Europe and aimed to identify determinants of inadequate treatment and variables associated with 30-day mortality by including all adult patients with S. aureus bacteraemia in 60 randomly selected hospitals over a period of two months. 21% of methicillin-susceptible S. aureus [MSSA] bacteraemias received inadequate empirical therapy, versus 28% of MRSA bacteraemias. Neither inadequate empirical treatment nor methicillin resistance was associated with 30-day mortality. We aimed to quantify the risk of mortality associated with MRSA versus MSSA bacteraemia, and to determine the effect of confounding factors hereon in a meta-analysis of observational studies. Comorbidity, severity of illness at onset of bacteraemia and appropriateness of treatment were considered the most important confounders. In 73 studies the unadjusted attributable mortality risk for mortality due to a bacteraemia caused by a MRSA versus MSSA was 13%. In 35 studies that adjusted for confounders the adjusted attributable mortality risk decreased to 9%. Finally, 7 studies that adjusted for the three key confounding variables yielded an attributable mortality risk of 4%.