The predictive value of NT-proBNP/ALB ratio for in-hospital pericardial effusion in young acute myocardial infarction patients

Figures

• 

  Figure 1. Flowchart of patient enrollment and study population selection. From January 2018 to October 2025, 19084 adult AMI patients were enrolled from the First Affiliated Hospital of Xi'an Jiaotong University. Among them, 2036 patients were ≤ 45 years old. After excluding 62 patients due to severe hepatic and renal insufficiency (n = 24), myeloproliferative disease (n = 1), and lack of liver function data or complete clinical data (n = 37), 1974 patients were finally included in the study analysis. These patients were divided into a Non-PE group (n = 1833) and a PE group (n = 141).

• 

  Figure 2. Forest plot of univariate logistic regression analysis for risk factors associated with in-hospital pericardial effusion Note: The forest plot displays the odds ratios (ORs) and 95% confidence intervals (CIs) for selected significant predictors identified in the univariate analysis. Variables are categorized into demographics, clinical characteristics, and laboratory findings. Abbreviations: BMI: body mass index; T2DM: type 2 diabetes mellitus; LVEF: left ventricular ejection fraction; STEMI: ST-segment elevation myocardial infarction; MI: myocardial infarction; BNP: B-type natriuretic peptide; ALB: albumin; cTnT: cardiac troponin T; HGB: hemoglobin; WBC: white blood cell; NEU: neutrophil; LYM: lymphocyte; MONO: monocyte; HCT: hematocrit; RDW: red blood cell distribution width; ALT: alanine aminotransferase; A/G: albumin to globulin; BUN: blood urea nitrogen; eGFR: estimated glomerular filtration rate; HbA1c: glycated hemoglobin; FT3: free triiodothyronine; TG: triglyceride; ApoA: apolipoprotein A; LpA: lipoprotein(a); hs-CRP: high-sensitivity C-reactive protein; PCT: procalcitonin; FIB: fibrinogen; INR: international normalized ratio; OR: odds ratio; CI: confidence interval.

• 

  Figure 3. Comparison of the discriminative performance between the baseline and the full predictive models (A) Receiver Operating Characteristic (ROC) curves for the prediction of in-hospital pericardial effusion in young AMI patients. Blue Line (Model 1): The baseline clinical model including Age, Killip class, STEMI diagnosis, BMI, Sex, and hs-CRP, which yielded an AUC of 0.718 (sensitivity: 0.670, specificity: 0.699, Youden index: 0.369 at the optimal probability cut-off of 0.0728). Red Line (Model 2): The full integrated model derived by adding lg (NT-proBNP/Albumin ratio), Plateletcrit (PCT), eGFR, Total Cholesterol, and LVEF to the baseline model. The full model demonstrated a significantly higher AUC compared to the baseline model (0.849 vs. 0.718, P < 0.001 by DeLong test), indicating superior predictive accuracy. Furthermore, at its optimal cut-off of 0.0729, Model 2 achieved an improved sensitivity of 0.777, and a Youden index of 0.554. Abbreviations: AUC, Area Under the Curve; BMI, Body Mass Index; eGFR, estimated Glomerular Filtration Rate; LVEF, Left Ventricular Ejection Fraction; PCT, Plateletcrit.