Combined Serum NGAL and Fetuin-A to Predict 28-Day Mortality in Patients with Sepsis and Risk Prediction Model Construction

Abstract

It was to investigate the predictive value of NGAL and Fetuin-A for 28-day mortality in patients with sepsis, and to construct a mortality risk prediction model. 120 patients admitted to The Affiliated Hospital of Xuzhou Medical University Hospital were grouped. Serum biochemical parameters were measured and scale scores were performed. The patient data were divided into a training set and test set in a ratio of 7:3, and the logistic regression model and random forest model were included to evaluate the 28-day mortality prediction efficacy of each index and model. The results showed that WBC, PLT, RBCV, and PLR decreased, SCr, Lac, PCT, D-dimer, NPR, NGAL, and Fetuin-A increased, APACHE II scale, SOFA scale, and OASIS scale scores increased in the death group (P < 0.05). SCr ≥ 408 μmol/L, Lac ≥ 2.3 mmol/L, PCT ≥ 30 ng/mL, D-dimer ≥ 2.33 mg/L, PLR ≥ 190, APACHE II ≥ 18 points, SOFA ≥ 2, OASIS ≥ 30, NGAL ≥ 352 mg/L, and Fetuin-A ≥ 0.32 g/L were found to be risk factors for 28-day death, while WBC ≥ 12 × 109/L, PLT ≥ 172 × 103/μL, and RBCV ≥ 30% were found to be protective factors for 28-day mortality. The predicted AUCs of APACHE II, SOFA, OASIS, NGAL, Fetuin-A, NGAL & Fetuin-A, logistic regression model, and random forest model were 0.80, 0.71, 0.77, 0.69, 0.86, 0.92, 0.83, and 0.81. NGAL combined with Fetuin-A has good prediction efficacy in 28-day mortality in septic patients.