176 Chapter 8 recruited adults (≥18 years) with monomicrobial methicillin-susceptible S. aureus (MSSA) bacteraemia in 19 Spanish University hospitals between 31/05/2019 and 24/02/2022, and randomised participants to receive cloxacillin (2g 6x/day) plus fosfomycin (3g 4x/day), or cloxacillin alone, for the initial seven days of treatment. Exclusion criteria included Child-Pugh class C liver cirrhosis, moderate-severe heart failure, injection drug use (IDU), MRSA bacteraemia, penicillin allergy, and acute SARS-CoV-2 infection. Ethical approvals were obtained from the South East Scotland Research Ethics Committee 02 (23/SS/0025) for the Edinburgh cohort study, the London (Westminster) Research Ethics Committee (12/LO/0637) for the ARREST trial, and the Spanish Medicines and Healthcare Products Regulatory Agency (AEMPS; 18-0905) and the Bellvitge University Hospital Ethics Committee (AC069/18) for the SAFO trial. Variables and definitions Comorbidities were defined according to the Charlson Comorbidity Index. Acquisition of infection was categorised according to the definitions used by Friedman and colleagues18. The source of infection was the most likely portal of entry of S. aureus into the bloodstream. Metastatic infection was defined as the presence of foci of infection remote from the portal of entry thought to have arisen through haematogenous dissemination. All-cause 84-day mortality was recorded in all cohorts. In the Edinburgh cohort, persistent bacteraemia was defined as a further positive blood culture during treatment >96h after the index blood culture and recurrent bacteraemia was defined as a further positive blood culture with the same S. aureus spa type within 90 days of stopping treatment3. In the ARREST cohort, microbiologic failure was defined as ongoing signs and symptoms of infection and growth of S. aureus from blood or a sterile site for >14 days from randomisation. Recurrence was defined as growth of S. aureus from a sterile site after >7 days of apparent clinical improvement. These were combined into a composite microbiologic outcome referred to as composite microbiologic failure. In the SAFO trial, persistent bacteraemia was documented at days 3 and 7 after randomisation. Statistical analyses Latent class analysis (LCA) was used to look for homogenous sub-groups within the larger heterogeneous cohorts of SAB using indicator variables selected based on availability and potential clinical relevance (consensus opinion of CDR, MS, ACW and GHG)19. Baseline patient and microbiologic variables were considered as class-defining variables which were first determined using data from the Edinburgh cohort, then this model was applied to the ARREST and SAFO cohorts. The classes were formed without any consideration of clinical or microbiological outcomes. We excluded variables with >10% missing data, categorical variables with >50% colinearity, any variable with <10% positivity unless considered of high clinical relevance, and any variable contributing <0·5% to the clustering20. Non-normally distributed values were log transformed for the LCA. Cases with missing values were handled with full information maximum likelihood (FIML), which is generally the preferred
RkJQdWJsaXNoZXIy MTk4NDMw