Background An objective topological analysis for any arbitrary network was provided by graph theory, which was implemented to reveal topological properties of brain networks in a wide range of neurodegenerative diseases. Objectives To investigate the difference of the topology of speech graphs among parkinsonian-type multiple system atrophy patients (MSA-P), idiopathic Parkinson’s disease (PD) patients, and healthy controls (HC). To explore the effect of levodopa on the topology of speech graphs in MSA-P and PD patients. Methods We applied a graph analysis of topology of speech graphs in MSA-P patients ( N = 39), idiopathic PD patients ( N = 51), and HC ( N = 51). All participants were assessed by a well-established semantic fluency test (animal). Participants’ verbal reactions were listed and represented as speech graphs with directionality. Results MSA-P patients mentioned more about their memory and remaining time than PD patients ( p 0.001) and HC ( p 0.001). Moreover, MSA-P patients’ speech graphs exhibited fewer nodes (PD: p = 0.004; HC: p 0.001), higher density (PD: p = 0.007; HC: p 0.001), smaller diameter (PD: p = 0.004; HC: p 0.001), and smaller average shortest path (PD: p = 0.002; HC: p 0.001) than both PD patients and HC. Compared with PD patients and healthy adults, the MSA-P patients generated denser and smaller speech graphs. Importantly, in MSA-P, the positive correlations between levodopa equivalent daily dose and diameter ( r = 0.39, p = 0.017) and average shortest path ( r = 0.38, p = 0.017) were obtained, suggesting that higher levodopa doses were associated with producing sparser and larger speech graphs in MSA-P patients. However, there was no significant correlation between levodopa equivalent daily dose and graph indices in PD patients. Conclusion MSA-P patients’ speech graphs were denser and smaller than both PD patients and healthy adults. The graph analysis of semantic fluency can differentiate MSA-P from PD to some extent. Importantly, in MSA-P patients, the impaired speech topology can be partially restored by levodopa.
Liu et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: