Risk of hospitalization and death within 2years after methicillin-resistant Staphylococcus aureus (MRSA) diagnosis in persons colonized or infected with livestock and non-livestock-associated MRSA-A nationwide register-based cohort study.

Abstract

Livestock-associated methicillin-resistant Staphylococcus aureus (LA-MRSA) CC398 has emerged in humans throughout Europe and the USA during the last decade and is associated with the spread of LA-MRSA CC398 in production animals. In this study, we investigated the risk of subsequent hospitalization with an S.aureus-related diagnosis and death within the first 2years after MRSA diagnosis. The study included 7,521 carriers of MRSA, an age-matched reference population of 376,041 individuals and 7,607 patients infected with MRSA. Hazard ratios (HR) for hospitalization with an S.aureus-related diagnosis were 4.09 (95% CI: 2.78-6.00) and for death 1.21 (95% CI: 0.80-1.83) in LA-MRSA CC398 carriers compared with the reference population. Comparing carriers of LA-MRSA CC398 and non-CC398 MRSA, HR for hospitalization was 0.61 (95% CI: 0.37-0.99) and death 0.25 (95% CI: 0.16-0.40), respectively. Patients initially diagnosed with LA-MRSA CC398 or non-CC398 MRSA infection differed from MRSA carriers in terms of older age, higher Charlson comorbidity index score and longer hospital stays. HR for subsequent hospitalization and death was similar regardless having infection with LA-MRSA CC398 or non-CC398 MRSA at the time of MRSA diagnosis. We established that MRSA CC398 carriers have a lower risk of hospitalization and death up to 2years after MRSA diagnosis than non-CC398 MRSA carriers do. In contrast, MRSA carriage, regardless the MRSA strain is a burden in terms of hospitalization and death when compared to the background reference population. Further, we established that there are no differences in risk of subsequent hospitalization and death in patients initially diagnosed with MRSA CC398 or non-CC398 MRSA infection. We suggest that public health surveillance of MRSA clearly needs to distinguish between carriage and infection as well as strain type before any inference from number of cases to disease burden is made.