Multiple-scattering effects can severely influences ground-based lidar measurements when the optical depth is not negligible such as in presence of fog or clouds. This problem can be faced both analytically and by Monte Carlo methods, although as usual the analytical techniques require several simplifications about the microphysical mechanisms whereas Monte Carlo simulation constitutes a more direct approach. In this paper, we discuss an iterative Monte Carlo method to simulate photon multiple scattering in optically dense media. Our results show that it is possible to correct for the multiple scattering influence both extinction and backscattering coefficients obtained by Raman lidar. In particular, for the typical cirrus cloud, the presence of the multiple scattering can lead to an underestimation of the extinction coefficient as large as 100% whereas the backscattering coefficient is almost unaffected by such process. This in turn evidences the strong dependence of the lidar ratio on the multiple-scattering effect.