Development and Prospective Validation of an AI Model for Prognosis in ITP Patients Undergoing Coronary Revascularization

1-month incidence of Major Adverse Cardiovascular Events (MACE)

MACE is a composite endpoint defined as the occurrence of any of the following: all-cause mortality, non-fatal myocardial infarction \[MI\], urgent coronary revascularization \[CRV\] and ischemic stroke. The time frame for assessment is from the date of CAD diagnosis (index date) until 1 month of follow-up.

1-year incidence of Major Adverse Cardiovascular Events (MACE)

MACE is a composite endpoint defined as the occurrence of any of the following: all-cause mortality, non-fatal myocardial infarction \[MI\], urgent coronary revascularization \[CRV\] and ischemic stroke. The time frame for assessment is from the date of diagnosis of CAD (index date) until 1 year of follow-up.

key predictors of adverse outcomes following revascularization

Identification of independent predictors for MACE in the revascularization group using a multivariate Cox proportional hazards regression model with stepwise selection or Lasso regularization. The model will include candidate clinical variables such as platelet count, type of CAD, comorbidities, and medication use. For each final predictor selected, the Hazard Ratio (HR), 95% confidence interval, and p-value will be reported. MACE is defined as a composite of all-cause death, non-fatal myocardial infarction, urgent coronary revascularization and ischemic stroke.

BARC type ≥2 bleeding event

clinical-related bleeding with a BARC type ≥2 bleeding event

overall bleeding event

the bleeding event identified according to the BARC standardized bleeding Criteria

Hospitalization for CAD within 1 year.

Hospitalization for CAD within 1 year.

Performance of the AI-based model in predicting Major Adverse Cardiovascular Events (MACE)

Assessment of the discriminative power and calibration of the AI-based prognostic model for predicting MACE in the validation cohort. Discrimination will be quantified using the Area Under the Receiver Operating Characteristic Curve (AUC) or C-statistic. Calibration will be assessed using a calibration plot and the Hosmer-Lemeshow goodness-of-fit test.MACE is defined as a composite of all-cause death, non-fatal myocardial infarction, urgent coronary revascularization and ischemic stroke.

Subgroup analysis: Impact of clinical factors on MACE incidence

Incidence rates (number and proportion of events) of MACE will be calculated for subgroups stratified by pre-defined clinical factors. These factors include type of Coronary Artery Disease (Stable CAD vs. Acute Coronary Syndrome \[ACS\]), platelet count categories (e.g., \<25×10⁹/L, 25-50×10⁹/L, \>50×10⁹/L), age, and sex. Furthermore, the association between these factors and MACE will be evaluated using multivariate Cox proportional hazards or logistic regression models, reported as adjusted Hazard Ratios (HR) with 95% confidence intervals. MACE is defined as a composite of all-cause death, non-fatal myocardial infarction, urgent coronary revascularization and ischemic stroke.