Monte Carlo (MC) modeling of a 6 MV photon beam was used to study the dose perturbation from a titanium rod 5 mm in diameter in various small fields range from 2 × 2 to 5 × 5 cm(2). The results showed that the rod increased the dose to water by ∼6% at the water-rod interface because of electron backscattering and decreased the dose by ∼7% in the shadow of the rod because of photon attenuation. The Pinnacle(3) treatment planning system calculations matched the MC results at the depths more than 1 cm past the rod when the correct titanium density of 4.5 g cm(-3) was used, but significantly underestimated the backscattering dose at the water-rod interface. A CT-density table with a top density of 1.82 g cm(-3) (cortical bone) is a practical way to reduce the dosimetric error from the artifacts by preventing high density assignment to them, but can underestimates the attenuation by the titanium rod by 6%. However, when multi-beam with intensity modulation is used in actual patient spinal stereotactic radiosurgery treatment, the dosimetric effect of assigning 4.5 instead of 1.82 g cm(-3) to titanium implants is complicated. It ranged from minimal effect to 2% dose difference affecting 15% target volume in the study. When hardware is in the beam path, density override to the titanium hardware is recommended.