Nomogram for Predicting Hemorrhagic Transformation Risk in Acute Ischemic Stroke Patients With Atrial Fibrillation

Background

Hemorrhagic transformation (HT) is a critical complication in acute ischemic stroke (AIS) patients with atrial fibrillation (AF) awaiting anticoagulation reinitiation. No reliable predictive model exists for assessing HT risk for these patients. Clinical decisions typically rely on NIHSS score and infarct size; however, other relevant risk factors remain insufficiently explored. This study aimed to develop and validate a predictive model for assessing the risk of HT in AIS patients with AF from stroke onset to anticoagulation therapy reinitiation.

Methods

This retrospective study included AIS patients with AF from two comprehensive medical centers in China. The primary outcome was HT postinfarction confirmed with CT/MRI before anticoagulation reinitiation. Significant predictors were identified via LASSO regression in the training set, followed by multivariable logistic regression for developing a predictive model and generating the nomogram. Model performance was validated in a separate external cohort.

Results

In the training cohort (n = 629), 174 patients (27.7%) developed HT. LASSO logistic regression revealed that infarct size, NIHSS score, diabetes mellitus, reperfusion therapy, left ventricular ejection fraction, and prehospital antihypertensive treatment were significant HT predictors. In the external validation cohort (n = 236), 61 patients (25.8%) developed HT. The nomogram exhibited strong predictive performance, with AUCs of 0.720 in the training set and 0.747 in the validation set.

Conclusions

The proposed nomogram offers a practical tool for predicting HT risk in AIS patients with AF before anticoagulation reinitiation, potentially supporting informed clinical decision-making, though further validation is required.