Prediction of early hematoma enlargement in hypertensive intracerebral hemorrhage using CT and CTA imaging features.

Abstract

Objective: To explore the relationship between CT low density, CTA dot and early hematoma enlargement in patients with hypertensive intracerebral hemorrhage (HICH), to analyse the influencing factors of hematoma enlargement and construct the prediction model.

Methods: A retrospective analysis was performed on the clinical data of 400 patients with HICH in the hospital between August 2022 and July 2024, and patients were randomly divided into the modeling group (n = 300) and the verification group (n = 100). According to examination results of head CT at admission, disease deterioration or 24 h after admission, hematoma volume was calculated by Tada formula. The increment of hematoma volume > 33% was defined as hematoma enlargement. The patients in the modeling group were divided into hematoma enlargement group and non-enlargement group. The relationship between CT, CTA signs and early hematoma enlargement was analyzed by Logistic regression analysis.

Results: The patients in the modeling group were divided into hematoma enlargement group (92 cases) and non-enlargement group (208 cases). The results of logistic analysis showed that large baseline hematoma volume, irregular morphology, island sign, low density sign, mixed sign, black hole sign and CTA spot sign were independent risk factors of early hematoma enlargement in HICH patients. According to data verification in modeling group, area under ROC curve (AUC), sensitivity and specificity were 0.817 [95%CI (0.768-0.866)], 78.26% and 73.56%, respectively. According to the principle of maximum Youden index, the cut-off value was 0.302.

Conclusion: The risk factors of early hematoma enlargement are analyzed and screened in HICH patients to construct a prediction model. The model is proved to have good validity, which can provide reference for the identification of high-risk groups.