Brain waste-clearance signals run low in depression, imaging study finds

People with major depressive disorder show weaker coupling between brain blood-oxygen signals and cerebrospinal fluid signals, an imaging measure used as a proxy for glymphatic system function, according to a study published in BMC Psychiatry. The alterations appeared at the whole-brain level and, most distinctly, within the dorsal attention network, one of the brain’s large-scale functional networks. The work, led by Hongfang Li, Fengxia Zhang, Yuhan Yang and colleagues with Aihong Yu as senior author, used a 5 Tesla MRI system to compare 32 patients with MDD against 35 healthy controls.

The glymphatic system is the brain’s waste-clearance network, in which cerebrospinal fluid exchanges with interstitial fluid along channels surrounding blood vessels. Because this process is difficult to observe directly in living humans, researchers rely on indirect imaging proxies. One proxy is the volume of the choroid plexus, the tissue that produces cerebrospinal fluid. Another is the DTI-ALPS index, derived from diffusion tensor imaging along the perivascular space. The third, and the one that carried the strongest signal in this study, is the coupling between blood-oxygen-level-dependent signals and cerebrospinal fluid signals, which captures a physiological interplay linked to glymphatic activity.

At the global level, patients with MDD showed significantly weaker gBOLD-CSF coupling than healthy controls, with a Cohen’s d of 0.843, a 95 percent confidence interval of 0.179 to 1.339, and a p value of 0.003. By contrast, the other two proxies showed no significant between-group differences: choroid plexus volume and the DTI-ALPS index were statistically indistinguishable between patients and controls.

The researchers then divided the brain into seven functional networks and calculated BOLD-CSF coupling within each. After Bonferroni correction for multiple comparisons, only one network survived: the dorsal attention network. Patients with MDD showed weaker coupling in this network, with a Cohen’s d of 0.835, a 95 percent confidence interval of 0.208 to 1.334, and a p value of 0.019. The specificity points to a network-level rather than a uniform whole-brain alteration.

Within the patient group, the coupling measures also tracked clinical severity. Depression was rated using the 17-item Hamilton Depression Rating Scale. Global gBOLD-CSF coupling was negatively correlated with the suicide item of the scale, meaning that patients with weaker coupling tended to score higher on this item. The correlation was ρ = −0.475, with a 95 percent confidence interval of −0.708 to −0.111, a p value of 0.011, and an FDR-adjusted q of 0.028. Network-specific coupling in the dorsal attention network showed a negative association with the same item, at ρ = −0.556, with a 95 percent confidence interval of −0.769 to −0.200, a p value of 0.002, and an FDR-adjusted q of 0.017.

The dorsal attention network finding extended to brain structure as well. Within the MDD group, network-specific BOLD-CSF coupling in this network was negatively associated with gray matter volume, with r = −0.504, a 95 percent confidence interval of −0.739 to −0.162, a p value of 0.006, and an FDR-adjusted q of 0.025. In other words, patients whose coupling was weaker tended to have lower gray matter volume. The authors interpret this as evidence that the coupling alterations are linked to both clinical symptoms and structural brain measures in depression.

These correlations are cross-sectional and associational, and the study cannot establish that impaired glymphatic function causes depressive symptoms or gray matter loss, or the reverse. The sample size, 32 patients and 35 controls, is modest, and the confidence intervals for the effect sizes are wide. The findings should therefore be read as evidence of altered glymphatic-related imaging measures in MDD, particularly within the dorsal attention network, and as a basis for further work rather than as a definitive characterization of the disorder.

The clinical relevance of the suicide-item correlations warrants careful interpretation. The correlations are with a single item on a rating scale, not with suicide attempts or deaths, and the study did not examine prospective outcomes. The authors note that major depressive disorder is associated with neuroinflammation, which may impair glymphatic system function, framing the study’s motivation.

Published on 3 November 2026 as an open-access article, the study adds network-level detail to a question that has mostly been examined at the whole-brain scale. Whether the dorsal attention network alterations prove to be a stable biomarker, a state marker that fluctuates with symptom severity, or a consequence of other disease processes remains to be determined by longitudinal and larger-sample studies. For now, the work documents that a functional imaging signature tied to glymphatic activity differs between people with major depressive disorder and healthy controls, and that within the patient group it aligns with symptom severity and gray matter volume.

Subject of Research: Glymphatic-related imaging alterations in major depressive disorder measured with multimodal MRI

Article Title: Global and network-specific glymphatic-related alterations in major depressive disorder: evidence from multimodal imaging

Article References: Li, H., Zhang, F., Yang, Y., Zhang, Y., Li, X., Li, E., Long, F., Yu, D., Zhou, J., Wang, Y., Zhou, Y., Ji, X., Jia, X., Wang, G., & Yu, A. (2026). Global and network-specific glymphatic-related alterations in major depressive disorder: evidence from multimodal imaging. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08825-0

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08825-0

Keywords: major depressive disorder, glymphatic system, BOLD-CSF coupling, dorsal attention network, DTI-ALPS index, choroid plexus volume, 5 Tesla MRI, Hamilton Depression Rating Scale, gray matter volume, neuroinflammation, BMC Psychiatry, multimodal imaging