Work

The ipsilesional DTI-ALPS index provides modest incremental prognostic value beyond clinical models for 90-day outcome in patients with acute MCA occlusion treated with intravenous thrombolysis: a prospective exploratory cohort study

Year: 2026

Type: article

Abstract: Background and purpose Impaired perivascular fluid dynamics may contribute to secondary brain injury after acute ischemic stroke, yet its prognostic relevance remains uncertain. We investigated whether the Diffusion Tensor Image Analysis along the Perivascular Space (ALPS) index—a quantitative MRI marker of perivascular fluid dynamics—provides incremental predictive value for long-term functional outcomes in patients with acute middle cerebral artery occlusion (MCAO) treated with intravenous thrombolysis. Methods In this prospective observational cohort study, 128 patients underwent multimodal brain MRI within 24 h of admission, and the ALPS index was calculated in the hemisphere ipsilateral to the occlusion (ipsilesional ALPS). Favorable outcome at 90 days was defined as modified Rankin Scale (mRS) 0–2. Multivariable logistic regression identified independent predictors, with covariates (admission National Institutes of Health Stroke Scale [NIHSS] score, LDL cholesterol, and smoking status) prespecified on clinical grounds rather than selected by univariate screening. Incremental discriminative value was assessed by comparing nested models using receiver operating characteristic (ROC) analysis and the likelihood-ratio test (LRT), which is preferable to DeLong’s test for nested models. Calibration (Hosmer–Lemeshow test, Nagelkerke R 2 ) and internal validation with 1,000 bootstrap resamples (optimism-corrected AUC) were performed. Interpretation Lower admission NIHSS score [adjusted odds ratio (aOR) per 1-point increase: 0.890, 95% CI: 0.813–0.976; p = 0.013] and lower LDL cholesterol (aOR per 1 mmol/L: 0.660, 95% CI: 0.472–0.923; p = 0.015) were independently associated with favorable outcome, whereas the ipsilesional ALPS index showed a directionally consistent, non-significant association (aOR per 0.1-unit increase: 1.525, 95% CI: 0.948–2.454; p = 0.082). Adding the ALPS index to a base model containing NIHSS and LDL yielded a modest but significant improvement in discrimination (AUC 0.681 to 0.699; ΔAUC = 0.018; LRT p = 0.042); calibration was adequate (Hosmer–Lemeshow p = 0.31 and 0.28), and the gain persisted after bootstrap optimism correction (ΔAUC = 0.013). Conclusion In this exploratory cohort, the ipsilesional ALPS index showed a consistent directional association with 90-day functional outcome and provided a modest but significant incremental improvement in discrimination beyond NIHSS and LDL cholesterol; the magnitude of this gain is small and requires external validation.

Source: Frontiers in Aging Neuroscience

DOI: 10.3389/fnagi.2026.1917297