Clinical and laboratory predictors of mortality in Staphylococcus aureus bacteremia in a high-risk setting: a single-center retrospective analysis of Pitt score, SOFA, neutrophil-to-lymphocyte ratio, and platelet-to-lymphocyte ratio.

OBJECTIVE: Staphylococcus aureus bacteremia (SAB) is associated with significant morbidity and mortality despite advancements in antimicrobial therapies. Timely risk stratification is critical to optimize patient outcomes. This study aimed to evaluate the prognostic value of clinical (Pitt bacteremia score [PBS]), combined clinical and laboratory (Sequential Organ Failure Assessment [SOFA]), and laboratory-based parameters (neutrophil-to-lymphocyte ratio [NLR] and platelet-to-lymphocyte ratio [PLR]) in patients with SAB.

METHODS: We conducted a retrospective observational study of 150 adult patients diagnosed with monomicrobial SAB between November 2023 and November 2024 at a tertiary care center. The predictive value of PBS, SOFA, NLR, and PLR for in-hospital mortality was assessed using ROC analysis and logistic regression.

RESULTS: The overall in-hospital mortality rate was 46%. PBS >4 and NLR >21 were independently associated with increased mortality (OR 11.72 and 6.36, respectively; p < 0.001). Absence of fever (OR 0.11) and presence of central venous catheter (OR 4.11) were also significant predictors. ROC analysis demonstrated good predictive performance for both PBS (AUC: 0.82) and SOFA (AUC: 0.795) scores with no statistically significant difference between them. Community-acquired SAB and primary bacteremia were associated with worse prognosis. Methicillin resistance did not significantly affect mortality. Fever response was significantly blunted in older adults but not in immunosuppressed patients.

CONCLUSION: High PBS and NLR are strong and independent predictors of mortality in SAB. As this study was conducted in a high-risk cohort with a considerable proportion of critically ill and immunocompromised patients, the findings are particularly relevant to this vulnerable population. External validation in larger, multicenter cohorts is required to confirm their generalizability.