The widespread use of social platforms for rapid information dissemination has intensified the risks associated with disinformation attacks, undermining access to credible sources and jeopardizing social stability. In this context, it is imperative to refine current approaches and develop novel solutions aimed at safeguarding the digital environment. Most efforts to determine the veracity of disseminated content have centered on the analysis of linguistic, stylistic, and content-based features. In contrast, the propagation process itself - illustrated through propagation graphs - remains relatively underexplored, despite its potential to reveal critical differences between genuine and deceptive information. In the present study, we conduct an extensive statistical analysis of datasets derived from propagation graphs. Utilizing metrics such as node centrality measures and various network parameters, we uncover structural differences in the spread of true versus false information across diverse social media ecosystems. Our findings indicate that the veracity of news plays a crucial role in shaping its propagation dynamics and that platform-specific features further modulate these patterns. This research represents a pivotal preliminary step that lays the groundwork for future efforts. We envision that subsequent studies will develop advanced classifiers by employing multiple graph representations alongside sophisticated machine learning techniques, ultimately enhancing the generalization and robustness of disinformation detection across diverse digital platforms.