What question did this study set out to answer?

This study aims to synthesize imaging findings related to glymphatic alterations in individuals with type 2 diabetes mellitus (T2DM).

May 25, 2026Open Access

Glymphatic-related imaging findings in type 2 diabetes mellitus: a systematic review and exploratory meta-analysis of DTI-ALPS studies

Key Points

This study aims to synthesize imaging findings related to glymphatic alterations in individuals with type 2 diabetes mellitus (T2DM).
Conducted a systematic review and meta-analysis of DTI-ALPS studies comparing T2DM and healthy controls.
Applied random-effects models with Hartung–Knapp–Sidik–Jonkman adjustment to analyze data.
Performed subgroup analyses differentiating uncomplicated T2DM from T2DM with complications.
In uncomplicated T2DM (4 studies), pooled effect size was Hedges’ g = -0.89 [95% CI (-1.45, -0.34); p = 0.015].
In complicated T2DM (3 studies), pooled effect size was Hedges’ g = -1.34 [95% CI (-3.55, +0.86); p = 0.119].
Point estimates indicated lower DTI-ALPS indices in T2DM across all studies, with statistical significance in the uncomplicated subgroup only.

Abstract

Background Type 2 diabetes mellitus (T2DM) is associated with white matter microstructural, neurovascular, and cognitive changes. Perivascular fluid dynamics may also be altered in T2DM. The diffusion tensor image analysis along the perivascular space (DTI-ALPS) index is an indirect imaging measure of perivascular water diffusivity, but its behavior across different clinical phenotypes of T2DM has not been quantitatively synthesized. Methods We conducted a systematic review and meta-analysis of DTI-ALPS studies comparing individuals with T2DM and healthy controls. Random-effects models with Hartung–Knapp–Sidik–Jonkman (HKSJ) adjustment were applied. Subgroup analyses were performed according to clinical phenotype, distinguishing uncomplicated T2DM from T2DM with established complications. Results Seven studies comprising 532 participants met the inclusion criteria. In the uncomplicated T2DM subgroup ( k = 4), the pooled Hedges’ g was −0.89 95% CI (−1.45, −0.34), HKSJ-adjusted; p = 0.015; I 2 = 57.9%, reaching statistical significance. In the primary analysis of complicated T2DM ( k = 3), the pooled Hedges’ g was −1.34 95% CI (−3.55, +0.86), HKSJ-adjusted; p = 0.119; I 2 = 90.7%, and did not reach statistical significance. The formal between-subgroup test was not significant in the primary analysis ( Q between = 0.66, df = 1, p = 0.416). A post-hoc sensitivity analysis of the complicated subgroup including Hu 2025 ( k = 4) yielded a pooled Hedges’ g of −1.73 95% CI (−3.42, −0.04); p = 0.047, whereas the between-subgroup difference remained non-significant ( Q between = 2.16, df = 1, p = 0.141). Across all seven studies and both subgroups, point estimates consistently favored lower DTI-ALPS indices in T2DM, although certainty of evidence was rated as very low for both primary outcomes. Conclusion This meta-analysis showed a consistent trend toward lower DTI-ALPS indices in T2DM than in healthy controls. The uncomplicated subgroup reached statistical significance, whereas the complicated subgroup primary analysis did not. No significant between-subgroup difference was detected. These findings should be interpreted cautiously given the limited number of studies, substantial heterogeneity in the complicated subgroup, and the indirect nature of the DTI-ALPS metric. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420261303719 , PROSPERO CRD420261303719.

Ask AI

Mark Helpful

Bookmark

Relay

View Full Paper