This paper focuses on the issue of adaptive event-triggered leader-follower consensus for multi-agents systems with output constraints and dead-zone inputs. By introducing an advanced nonlinear mapping technique to obtain the unconstrained auxiliary variables of constrained system states, a new systems model without output constraints is constructed. Unlike existing schemes, the proposed strategy can be used in both constrained and unconstrained situations without requiring changes to the control structure. Moreover, a state estimator is constructed to observe the unavailable states. To conserve communication resources, an event-triggered rule with a dynamic threshold is designed to decrease superfluous information transmissions from the controller to the actuator. It is proven that all signals in closed-loop systems are ultimately bounded and the system output does not violate the given constraint range. At last, a numerical simulation example is provided to confirm the correctness and efficiency of the proposed method.