Comment on: Competing Interactions, the Renormalization Group, and the Isotropic-Nematic Phase Transition, by D. Barci and D. Stariolo, Phys. Rev. Lett. 98, 200604 (2007)

In a recent PRL Barci and Stariolo (BS) generalized the well known Brazovskii model to include an additional rotationally invariant quartic interaction and study this model in two dimensions (2d). Depending on the parameters of the model, BS find two transitions: a first order isotropic-lamellar (striped) or a second order isotropic-nematic (which they speculate to be in the Kosterlitz-Thouless universality class). Using a simple symmetry argument, I show that the striped phase found by BS can not exist in 2d. Furthermore, I argue that based on the coarse-grained action used by BS it is impossible to reach any conclusion about the nature of the isotropic-nematic transition.