Do conventional intracerebral hemorrhage functional scores apply to mild-to-moderate hemorrhage?

Conventional intracerebral hemorrhage (ICH) functional scores are typically developed based on clinical factors for assessing mortality risk, and their applicability in evaluating functional outcomes in patients with the specific hemorrhagic subtype of mild to moderate ICH(MTM-ICH) remains unclear. Therefore, we aimed to construct a nomogram model to predict early functional dependence in this patient population and to validate three previous general ICH functional scores.

We conducted a retrospective analysis of data from 575 patients with MTM-ICH (GCS score ≥ 9) treated at our hospital between January 2018 and July 2025. The baseline demographic information, clinical characteristics, and clinical outcomes were collected. Early functional dependence was defined as a modified Rankin Scale (MRS) score ≥ 3 at discharge or 30 days after ICH. Least absolute shrinkage and selection operator (LASSO) and stepwise logistic regression analyses were used to screen for risk factors for early functional dependence and construct a predictive model. The model was visualized using a nomogram, and its predictive performance was compared with three existing ICH functional scores (FUNC, ICH-GS and ICH-FOS). Internal validation of the nomogram was performed using a 10-fold cross-validation combined with bootstrap resampling.

Among 575 patients with MTM-ICH, 198 (34.4%) developed early functional dependence. LASSO and stepwise logistic regression analyses identified six independent predictors: age, admission GCS score, hematoma volume, thalamic hemorrhage, basal ganglia hemorrhage, and mechanical ventilation. The nomogram based on these factors demonstrated good calibration and superior predictive performance compared to three existing ICH functional scores (AUC = 0.856 for our nomogram, 0.579 for FUNC, 0.601 for ICH-GS, and 0.714 for ICH-FOS, respectively). The decision curve analysis further demonstrated that the model had greater net benefit than the three previous general ICH functional scores. Internal validation confirmed the model's stability and generalizability (10-fold cross-validation mean: 0.846; bootstrap-corrected C-index: 0.844).

Among patients with MTM-ICH, advanced age, lower admission GCS score, larger hematoma volume, thalamic or basal ganglia hemorrhage, and mechanical ventilation were associated with a higher risk of early functional dependence. The nomogram developed based on these factors demonstrated high predictive performance. In contrast, conventional ICH functional scores demonstrate certain limitations when assessing this specific hemorrhage subtype.