<p>Epileptic seizure propagation is a central problem in understanding the mechanisms underlying epilepsy. Although extracellular potassium ([K⁺]_o) accumulation and synaptic network architecture have each been independently implicated in modulating seizure spread, a unified theoretical account of their interaction remains lacking. In this study, we develop a computational neuronal network model incorporating dynamic ion regulation to systematically examine the coupled effects of ionic gradients and synaptic topology on propagation dynamics. Results indicate that focal elevation in [K⁺]_o establishes spatial concentration gradients that elevate neuronal excitability and reduce action potential thresholds, thereby facilitating rapid propagation. Furthermore, such ionic perturbations amplify synaptic efficacy, promoting seizure diffusion. Network topology is found to critically constrain both the route and extent of propagation: under uniform [K⁺]_o distribution symmetrically structured networks with balanced connectivity constrain spread via inhibitory stabilization; in contrast, structural asymmetries predispose the network to excitation–inhibition (E-I) imbalance and global synchronization. Notably, elevated [K⁺]_o exhibits bidirectionality depend on inherent network inhibition: it enhances control in strongly inhibitory contexts, suppressing seizures, while further destabilizing E-I dynamics in weakly inhibitory networks and accelerating propagation. By integrating dynamic ion homeostasis with structural connectivity, this theoretical framework reveals the dual regulatory role of [K⁺]_o in shaping network excitability and E-I balance. The study provides a mechanistic basis for understanding seizure propagation and advances theoretical foundations for future modeling efforts.</p>