Development of a nomogram for predicting wall irregular pulsation and risk stratification in intracranial aneurysms

Accurate rupture risk stratification for unruptured intracranial aneurysms (UIAs) presents clinical challenges. Aneurysm wall irregular pulsation is an imaging biomarker of aneurysm vulnerability. This study aims to develop a nomogram for predicting wall irregular pulsation and risk stratification in intracranial aneurysms.

This was a consecutive cross-sectional observational study with prospective data collection for UIAs between 2017 and 2020. The irregular pulsation was defined as sequential morphological alterations observed in three or more consecutive phases by on four-dimensional CT angiography (4D-CTA). A generalized estimating equation (GEE) method identified independent predictors of the presence of wall irregular pulsation. A logistic regression-based predictive model was developed. Model performance was evaluated using receiver operating characteristic analysis.

We included 202 patients with 262 unruptured aneurysms. Multivariable analysis revealed four independent predictors of the presence of wall irregular pulsation: aneurysm location at ACA/PCom/Posterior (OR: 2.17, 95% CI 1.21-3.88; p = 0.0099), presence of daughter sac (OR: 5.15, 95% CI 1.91-13.86; p = 0.001), aspect ratio (OR: 1.80, 95% CI 1.09-2.99; p = 0.022), and neck size (OR: 1.23, 95% CI 1.02-1.48; p = 0.030). The developed model demonstrated good discrimination, achieving a sensitivity of 88.5%, specificity of 71.2%, overall accuracy of 76.9%, and an area under the curve (AUC) of 0.808 (95% CI, 0.704-0.913). Further validation using an independent cohort yielded comparable performance, with a sensitivity of 76.9%, specificity of 88.9%, overall accuracy of 81.8%, and an AUC of 0.855 (95% CI, 0.694-1.000). A nomogram was constructed to visualize the model.

This study presents a logistic regression-based model for predicting aneurysm wall irregular pulsation. The nomogram visually assesses individualized risks, demonstrating potential for future clinical implementation in bedside decision-making.