In this paper, high-order moments, even exponential moments, estimates are established for solutions to stochastic differential equations driven by fractional Brownian motion whose drifts are measurable and have linear growth. As applications, we study the weak uniqueness of solutions to fractional stochastic differential equations. Moreover, combining our estimates and the Fourier transform, we establish the existence of density of solutions to equations with irregular drifts. We further obtain Gaussian estimates for the tails of additive functional of such equation with bounded and measurable drift. It is relied on a mix of the Girsanov transform and the Clark-Ocone type covariance representation.