<p>This paper investigates an adaptive event-triggered distributed model predictive control (AETDMPC) problem for dynamically coupled nonlinear systems with input constraints and bounded additive disturbances. The overall system consists of multiple subsystems that exchange information with their neighbors via the communication network. Within this framework, mutual dynamic couplings and external disturbances exert a significant influence on subsystem performance. To restrict these effects, a consistency constraint is introduced to limit the deviation between the predicted and assumed subsystem states. Moreover, to enhance flexibility in alleviating computational and communication burdens, an AET mechanism (AETM) is developed, which regulates the frequencies of optimization updates and inter-subsystem communications through a state-dependent threshold that adjusts adaptively within predefined bounds. The consistency constraint combined with the AETM ensures the rigorous boundedness of the effects induced by couplings and disturbances, while guaranteeing Zeno-free phenomenon. Furthermore, sufficient conditions are derived to guarantee the algorithm iterative feasibility and closed-loop stability. Finally, simulation studies and comparative analysis validate the effectiveness of the AETDMPC strategy.</p>