An adpative integration technique for time advancement of particle motion in the context of coupled computational fluid dynamics (CFD) -discrete element method (DEM) simulations is presented in this work. CFD-DEM models provide an accurate description of multiphase physical systems where a granular phase exists in an underlying continuous medium. The time integration of the granular phase in these simulations present unique computational challenges due to large variations in time scales associated with particle collisions. The algorithm presented in this work uses a local time stepping approach to resolve collisional time scales for only a subset of particles that are in close proximity to potential collision partners, thereby resulting in substantial reduction of computational cost. This approach is observed to be 2-3X faster than traditional explicit methods for problems that involve both dense and dilute regions, while maintaining the same level of accuracy.