The paper discusses the principles of the Critical Shear Crack Theory (CSCT) in terms of the punching shear analysis of flat slabs made from lightweight aggregate concretes. The basic assumptions of the CSCT were discussed, explaining the differences with regard to the calculation of ordinary concrete flat slabs, relating mainly to the adopted failure criterion associated with ultimate slab rotation. Taking into account the observations and conclusions from the previous experimental investigations, it was confirmed, that contribution of lightweight aggregate particles in the aggregate interlock effect should be ignored, due to possibility of aggregate breaking. However, the analysis of the profile of failure surface confirmed, that particles of the natural fine aggregate increase the roughness of the surface and should be included by formulating failure criterion for LWAC slabs. The theoretical load-rotation relationships were compared with the results of measurements, confirming good agreement in most cases. The theoretical ultimate rotations were lower on average by about 11% than the experimental ones. The analysis of 57 results of the experimental investigations on punching shear of LWAC slabs made from various types of artificial aggregates showed a very good agreement with predictions of the CSCT. The obtained ratio of the experimental to theoretical load was 1.06 with a coefficient of variation of 9.1%. The performed parametric study demonstrated a low sensitivity of the correctness of the CSCT predictions to a change in a fairly wide range of parameters such as: the longitudinal reinforcement ratio, concrete compressive strength and concrete density.