Noninvasive assessment of glymphatic dysfunction in idiopathic normal pressure hydrocephalus with diffusion tensor imaging

J Neurosurg 140:612–620, 2024

Diffusion tensor imaging (DTI) along the perivascular space (ALPS) (DTI-ALPS)—by calculating the ALPS index, a ratio accentuating water diffusion in the perivascular space—has been proposed as a noninvasive, indirect MRI method for assessing glymphatic function. The main aim of this study was to investigate whether DTI-ALPS would reveal glymphatic dysfunction in idiopathic normal pressure hydrocephalus (iNPH) and whether the ALPS index was associated with disease severity.

METHODS Thirty iNPH patients (13 men; median age 77 years) and 27 healthy controls (10 men; median age 73 years) underwent MRI and clinical assessment with the Timed Up and Go test (TUG) and Mini-Mental State Examination (MMSE); only the patients were evaluated with the Hellström iNPH scale. MRI data were analyzed with the DTI-ALPS method and Radscale screening tool.

RESULTS: iNPH patients showed significantly lower mean ALPS index scores compared with healthy controls (median [interquartile range] 1.09 [1.00–1.15] vs 1.49 [1.36–1.59], p < 0.001). Female healthy controls showed significantly higher ALPS index scores than males in both hemispheres (e.g., right hemisphere 1.62 [1.47–1.67] vs 1.33 [1.14–1.41], p = 0.001). This sex difference was not seen in iNPH patients. The authors found a moderate exponential correlation between mean ALPS index score and motor function as measured with time required to complete TUG (r = −0.644, p < 0.001), number of steps to complete TUG (r = −0.571, p < 0.001), 10-m walk time (r = −0.637, p < 0.001), and 10-m walk steps (r = −0.588, p < 0.001). The authors also found a positive linear correlation between mean ALPS index score and MMSE score (r = 0.416, p = 0.001). Simple linear regression showed a significant effect of diagnosis (B = −0.39, p < 0.001, R 2 = 0.459), female sex (B = 0.232, p = 0.002, R 2 = 0.157), and Evans index (B = −4.151, p < 0.001, R 2 = 0.559) on ALPS index. Multiple linear regression, including diagnosis, sex, and Evans index score, showed a higher predictive value (R 2 = 0.626) than analysis of each of these factors alone.

CONCLUSIONS The ALPS index, which was significantly decreased in iNPH patients, could serve as a marker of disease severity, both clinically and in terms of neuroimaging. However, it is important to consider the significant influence of biological sex and ventriculomegaly on the ALPS index, which raises the question of whether the ALPS index solely reflects glymphatic function or if it also encompasses other types of injury. Future studies are needed to address potential confounding factors and further validate the ALPS method.