Clinical and Radiological Predictors for Early Hematoma Expansion After Spontaneous Intracerebral Hemorrhage: A...

Early hematoma expansion is a major determinant of poor outcomes after spontaneous intracerebral hemorrhage (ICH). Identifying reliable predictors of hematoma expansion may facilitate risk stratification and timely interventions. This study aimed to evaluate clinical, laboratory, and radiological factors associated with early hematoma expansion within 24 h.

We retrospectively analyzed consecutive patients with spontaneous ICH admitted to a tertiary hospital in Korea between 2009 and 2021. Inclusion criteria were aged ≥ 18 years, primary spontaneous ICH, baseline non-contrast CT (NCCT), and follow-up CT within 24 h. Clinical, laboratory, and medication histories were collected, and NCCT/CT angiography (CTA) imaging markers (spot sign, blend sign, hypodensity, swirl sign, black hole sign, island sign, mean hematoma density) were evaluated. Early hematoma expansion was defined as an absolute volume increase ≥6 cm³ or a relative increase ≥33% on follow-up CT. Multivariate logistic regression identified independent predictors.

Among 899 screened patients, 581 met inclusion criteria (mean age 61.6 years; 59.7% male). Seventy-eight patients (13.4%) experienced early hematoma expansion. Independent predictors included CTA spot sign (adjusted OR 9.001, 95% CI 4.414-18.354), blend sign (OR 3.054, 95% CI 1.349-6.910), mean hematoma density <60 HU (OR 2.432, 95% CI 1.271-4.655), male sex (OR 2.902, 95% CI 1.419-5.935), and statin use (OR 2.990, 95% CI 1.149-7.782). Prior antiplatelet therapy was associated with a reduced risk of hematoma expansion (OR 0.118, 95% CI 0.014-0.981).

Early hematoma expansion occurred in 13.4% of patients and was predicted by a combination of CTA and NCCT markers, as well as clinical and pharmacological factors. Spot sign remained the strongest predictor, while NCCT features such as blend sign and low hematoma density also provided practical prognostic value. These findings underscore the multifactorial pathophysiology of ICH expansion and highlight the importance of integrating imaging, clinical, and therapeutic variables into prediction models to improve early risk stratification and guide targeted interventions.