A Retrospective Cohort Study: Predictive Value of Biomarkers for Mortality in Bloodstream Infection Patients

Objective To investigate the predictive value of procalcitonin (PCT), uric acid, platelet count, and neutrophil-to-lymphocyte ratio (NLR) for mortality risk in blood culture-positive sepsis patients. Methods A retrospective analysis was conducted on 500 patients with positive blood cultures from both upper and lower limbs treated in the emergency department of our hospital from January 2023 to February 2025. Based on clinical outcomes, patients were divided into a rehabilitation discharge group (n = 416) and a death group (n = 84). Peripheral blood PCT, platelet count, uric acid, and NLR levels within 24 hours of admission were compared. Receiver operating characteristic (ROC) curves were used to evaluate the predictive value of these biomarkers for mortality. Binary logistic regression was employed to identify independent risk factors influencing mortality in blood culture-positive sepsis patients. Results Logistic regression analysis revealed that platelet count (OR = 0.996, 95% CI: 0.993–1.000), absolute lymphocyte count (OR = 0.901, 95% CI: 0.489–1.661), absolute neutrophil count (OR = 1.025, 95% CI: 0.976–1.077), PCT (OR = 1.012, 95% CI: 1.002–1.021), uric acid (OR = 1.010, 95% CI: 1.007–1.012), and NLR (OR = 1.040, 95% CI: 1.009–1.071) were significant risk factors for mortality ( p  < 0.05). ROC analysis demonstrated that AUC for platelet count, PCT, uric acid, NLR, and their combination were 0.622, 0.747, 0.759, 0.650, and 0.823, respectively, with the combined model showing the highest predictive value. Conclusion Elevated PCT, uric acid, and NLR levels, along with reduced platelet counts, are predictive of mortality in blood culture-positive sepsis patients. Clinicians should closely monitor these biomarkers for early intervention to reduce mortality.