Sepsis in Saudi Arabian acute care settings, a systematic review of epidemiology, prognostic markers, antimicrobial management, and prediction models

Abstract

Background: Sepsis is a primary driver of in-hospital mortality. Guideline driven bundles have not eliminated outcome variability in settings, and significant implementation challenges persist. Objectives: This review synthesizes Saudi and regional studies regarding sepsis epidemiology, prognostication, antimicrobial timing, stewardship, and the performance of prediction models in emergency and ICU settings.

Methods: We conducted a PRISMA aligned systematic review, searching electronic databases for adult studies in acute care hospitals. Eligible designs (observational cohorts, quality-improvement reports) were required to report mortality, ICU admission, organ support, length of stay, or model performance (discrimination, calibration). Two reviewers screened, extracted data, and assessed bias. Significant heterogeneity necessitated a narrative synthesis. Results: Ten studies met inclusion criteria. High ICU mortality from septic shock was a consistent finding. Both comorbidity and dynamic laboratory markers (platelet trajectories) had strong prognostic value. Evidence comparing antimicrobial timing (less than 1 hour versus 1-3 hours) in the ED was heterogeneous post adjustment. We also identified significant operational barriers to culture guided de escalation. Prediction models showed reasonable discrimination but variable calibration, requiring local validation before use.

Conclusions: Key gaps identified include strengthening stewardship, HAI prevention, and the cautious, validated-only use of predictive models. Priorities for future work include standardized definitions and prospective multicenter evaluations.