In a recent work [1], Bamba and Ogawa developed a microscopic model describing the field of a photonic cavity coupled to a matter exciton-like resonance. One of the results they obtain studying such a model is that, in the ultrastrong coupling regime, usually safe approaches can give wrong results for the dissipation rates of the polaritonic excitations. In particular the dissipation rates calculated applying the rotating wave approximation on the system-environment coupling qualitatively differ from the ones calculated using a microscopic theory based on the quantum electrodynamics for dielectric media. Here I show that this result is an artifact, caused by an inconsistent application of the rotating wave approximation and by a questionable parameter choice.In the light-matter ultrastrong coupling (USC) regime many usually safe approximations fail and it is thus necessary to pay great attention when studying this nonperturbative interaction regime. In particular, neglecting the light-matter interaction while calculating the systemenvironment coupling in the USC regime can lead to wrong or unphysical results [2][3][4].A simple way to study open quantum systems in the USC regime is to apply the rotating wave approximation (RWA) on the system-environment coupling in the dressed states basis. In this way positive and negative frequency excitations do not mix and the global ground state is the product of the systems and environments ground states [4,5]. Until now no work had addressed the important question of the reliability and generality of such an approach.In Ref.[1] the authors developed a microscopic model describing a photonic cavity filled with infinite-mass oscillators. Using Maxwell boundary conditions (MBC) they calculated the lifetimes of the cavity eigenmodes due to the finite reflectivity of one of the mirrors, as a function of the strength of the light-matter coupling. The microscopic calculation proves that the lack of positive and negative frequency mixing is not a simplification imposed by the RWA but it emerges naturally when one considers in a consistent way a microscopic theory based on the quantum electrodynamics for dielectric media. One of the consequences of such a theory is that in the USC regime the dissipation rates calculated using MBC differ qualitatively from the ones obtained using the usual system-environment RWA approach. This result is very surprising and, if true, it could have a major impact on future investigations on USC physics, as it would imply that in this regime the system observables can critically depend on microscopic details. The bottom panel of Fig. 2 in Ref.[1], replicated here in the panel (a) of Fig. 1, shows the dissipation rates for a resonant cavity mode, calculated using the MBC approach compared with the ones calculated using the RWA. They clearly differ in two main regards: (I) the values given by the RWA are up to one order of magnitude larger that the MBC ones and (II) the MBC gives larger losses for the lower polariton, while the RWA for the upper one.In ...