Establishment of a nomogram prediction model for intracranial hemorrhage risk after mechanical thrombectomy / 中国医师杂志

ABSTRACT

Objective:To establish a nomogram model for predicting the risk of intracranial hemorrhage in patients with acute cerebral infarction after mechanical thrombectomy.Methods:The clinical data of 251 patients with acute cerebral infarction who underwent mechanical thrombectomy in Shengjing Hospital Affiliated to China Medical University from January 2017 to December 2019 were retrospectively analyzed. Logistic regression model was used to analyze the independent risk factors of intracranial hemorrhage after mechanical thrombectomy. A nomogram prediction model based on independent risk factors was established to verify the prediction and accuracy of the model.Results:The analysis results of logistic regression model were as follows age ( OR=1.303, 95% CI1.184-1.433), the time from infarction to re-canalization ( OR=4.306, 95% CI2.497-7.425), preoperative NISS score ( OR=7.584, 95% CI2.221-25.900), preoperative computer tomography (CT) low-density lesions ( OR=7.954, 95% CI1.176-53.792) were independent risk factors for intracranial hemorrhage after mechanical thrombectomy in patients with acute cerebral infarction ( P<0.05). Based on the above 4 independent risk factors, a nomogram predictive model of intracranial hemorrhage risk after mechanical thrombectomy was established. The Bootstrap internal verification method proved that the model had good prediction accuracy, and the receiver operating characteristic (ROC) curve analysis testified that area under curve (AUC) area was 0.966. Conclusions:The risk nomogram prediction model has good accuracy, discrimination and good prediction ability, which can improve the diagnostic efficacy of intracranial hemorrhage after mechanical thrombectomy in patients with acute cerebral infarction.