In aircraft engine control, replacing linear regulators by sliding mode control (SMC) regulators is considered as an effective approach to reducing the conservatism in the traditional treatment for limit protection. However, most of the relevant studies are based on linear descriptions, which cannot represent the nonlinear systems directly due to their limited valid range. Even if gain scheduling techniques are employed, the stability of the nonlinear systems cannot be theoretically guaranteed. In this paper, a sliding mode strategy for a class of uncertain linear parameter varying (LPV) systems is studied. LPV descriptions are applied to extend the valid range of the linear models covering the entire operation envelope with guaranteed performance and stability. The mismatch between LPV and the real systems is considered as uncertainties. With a sliding surface defined by the tracking errors, system properties on the surface are proved to be satisfactory. After that, a reaching law is designed to ensure global invariance of SMC. Based on a reliable model turbofan, simulation results show that the SMC method can fully exploit the limit margin and, compared to the traditional proportional-integral-derivative (PID) control, has a faster response. In addition, stability and effectiveness of the proposed method are verified in a temperature protection case.