# A nomogram model to predict cognitive impairment in patients with spontaneous intracerebral hemorrhage

## TL;DR

This study creates a prediction model to identify patients at risk of cognitive decline after brain hemorrhage, using factors like infection, brain injury size, and education level.

## Contribution

A novel nomogram model integrating acute complications and sociodemographic factors to predict post-stroke cognitive impairment in sICH patients.

## Key findings

- Pulmonary infection and larger hematoma volume increase risk of cognitive impairment after sICH.
- Higher years of education are associated with reduced risk of cognitive decline.
- The nomogram model shows strong predictive accuracy with AUCs of 0.771 and 0.820 in training and validation sets.

## Abstract

Post-stroke cognitive impairment (PSCI) after spontaneous intracerebral hemorrhage (sICH) is highly prevalent and severely impacts patients’ long-term quality of life. However, accurate prediction tools that integrate acute-phase complications with sociodemographic characteristics are currently lacking. This study aimed to identify independent risk factors for PSCI in sICH patients and to construct a visual nomogram prediction model to guide clinical risk stratification prior to hospital discharge.

We retrospectively analyzed clinical data from 264 sICH patients admitted to the Affiliated Hospital of Xuzhou Medical University between July 2023 and July 2025. Patients were classified into cognitive impairment and cognitively normal groups based on the Montreal Cognitive Assessment (MoCA) score (<22). The dataset was randomly split into a training set (n = 198, 75%) and a validation set (n = 66, 25%). Univariate and multivariate logistic regression analyses were employed to screen for independent predictors, which were then used to construct the nomogram model. The model’s discriminative ability, calibration, and clinical utility were validated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA).

The overall incidence of PSCI in this cohort was 44.3%. Multivariate logistic regression analysis identified pulmonary infection (OR 3.980, 95% CI 2.075–7.635, p = 0.002) and hematoma volume (OR 1.030, 95% CI 1.015–1.045, p < 0.001) as independent risk factors for PSCI, whereas years of education (OR 0.885, 95% CI 0.831–0.944, p < 0.001) served as an independent factor associated with reduced risk. The nomogram model demonstrated excellent discriminative ability with AUCs of 0.771 and 0.820 in the training and validation sets, respectively. Calibration curves indicated high consistency between predicted probabilities and observed outcomes. DCA confirmed clinical net benefit across a wide range of threshold probabilities.

This study successfully developed a nomogram prediction model incorporating pulmonary infection, hematoma volume, and years of education. The model suggests that cognitive decline after sICH is associated with a combination of systemic inflammation (brain–body axis interaction), primary structural injury, and insufficient cognitive reserve. This user-friendly and accurate scoring tool can assist clinicians in identification of high-risk subgroups for PSCI upon completion of inpatient care, thereby informing intensified clinical monitoring and rehabilitation planning.