PREDICTORS OF MORTALITY IN MULTIDRUG-RESISTANT SEPSIS: A SYSTEMATIC REVIEW

Abstract

Introduction: Multidrug-resistant (MDR) bacterial infection complicating sepsis and septic shock is associated with substantial mortality. The excess risk is not determined by antimicrobial resistance alone; it reflects the interaction of organ dysfunction, septic shock, comorbidity, infection source, adequacy and timing of antimicrobial treatment, and source control. Aim: To systematically synthesise clinical, microbiological and treatment-related predictors of mortality in patients with sepsis or severe bloodstream infection caused by multidrug-resistant or carbapenem-resistant bacteria. Materials and Methods: A systematic review was conducted in accordance with PRISMA 2020 principles. A structured search of PubMed/MEDLINE and a multidisciplinary academic literature index was performed for studies available up to June 2026. Original observational studies enrolling patients with MDR, extensively drug-resistant or carbapenem-resistant bacterial sepsis, septic shock, bloodstream infection, or severe ICU infection were eligible when multivariable mortality predictors were reported. The targeted search set yielded 30 records; one duplicate was removed, 29 records were screened, 21 full-text reports were assessed, and 15 studies were included in the qualitative synthesis. Risk of bias was appraised using Joanna Briggs Institute (JBI)-informed domains for observational studies. Owing to major heterogeneity in organisms, outcomes and effect measures, meta-analysis was not performed. Results: Across the included studies, the most reproducible predictors of death were septic shock and greater acute illness severity, reflected by SOFA, APACHE II, SAPS II or Pitt bacteraemia scores. Inappropriate or inactive initial antimicrobial therapy repeatedly increased mortality, particularly in Gram-negative severe sepsis, carbapenem-resistant Enterobacterales and drug-resistant Acinetobacter infections. Other independent predictors included advanced age, greater comorbidity burden, malignancy or immunocompromised state, pneumonia or non-urinary sources, mechanical ventilation, high carbapenem minimum inhibitory concentration, and absence of adequate source control. Early appropriate therapy and demonstrable clinical improvement were consistently associated with improved survival. Conclusion: Mortality in MDR sepsis is driven by the convergence of host vulnerability, organ failure, shock, difficult to-treat pathogens and delays or inadequacy of effective treatment. Early recognition of high-risk patients, prompt active antimicrobial therapy, rapid source control and repeated severity assessment should be central to management and future risk-prediction models.