More insights into disruption and decoupling of individual metabolic connectomes in Parkinson's disease.

PURPOSE: Metabolic disturbances are hallmark pathological features of Parkinson's disease (PD) and can be noninvasively captured by arterial spin labeling (ASL). However, the metabolic pattern of disconnections beyond regional alterations remains scarcely documented. We aimed to comprehensively investigate metabolic impairments in PD at the individual network scale by utilizing abundant hemodynamic metrics.

METHODS

The multi-delay (m-ASL) was employed to obtain corrected cerebral blood flow (CBF), arterial cerebral blood volume and arterial transit time from groups of PD patients and healthy controls (HCs). For the comparison or cross-modality analysis, the uncorrected CBF and grey matter volume were also acquired via the ASL approach with single post-labeling delay (s-ASL) and T1 sequences, respectively. Metabolic similarity network (MSN) based on perfusion data and structural similarity network (SSN) were constructed for each individual, followed by analyses of connectivity, topology, classification and coupling.

RESULTS

Metabolic similarity networks in patients with PD belong to small-world connectomes but exhibit decreased global integration and local segregation at the global level. The impairments in nodal centralities, modular architectures and connectivity strengths were diverse among MSNs. m-ASL improved the diagnostic efficiency of CBF-MSN by the CBF correction and further optimized the classification performance via the integration of all MSNs. Decoupling of SSN-MSN presented pathologically increased coefficients within subnetworks, deriving from altered metabolic and structural connectivity.

CONCLUSION

Our study revealed the complex metabolic disconnections and SSN-MSN decoupling that underlie the complicated neurodegenerative process in PD, highlighting the clinical implications of m-ASL for comprehensive investigations of the metabolic pattern of disconnection syndromes.