RF heating causes most of incidents during MRI examinations. Since metallic implants are potential materials inducing heating by RF exposure in MRI, reporting the results of computer simulation of RF heating due to the products of metallic implants has been obliged. However, there still are patients who were implanted metallic products before the advent of MRI and MR compatibility of most such products is unknown. It was reported that an MRI examination of a patient implanted an intramedullary rod in his forearm was aborted due to a heating claim from the patient ¹. In this study, we confirm RF heating of such patient by using an electromagnetic analysis software dedicated for MRI, and shows that positioning of an implanted arm can decrease SAR sufficiently enough to take MRI examinations.The finite-difference time-domain software Sim4Life (ZMT, Zurich, Switzerland) was used for SAR simulation with the anatomically-based model Duke. The generic MRI birdcage model (60 cm length) operating at 64 MHz was used as a transmission RF coil. The intramedullary rod was modeled by a tube (2 cm diameter, 25 cm length) with a material of stainless steel (1.35×10⁶ S/m). To accelerate the calculation time, the small mesh size was employed for the forearm and the larger mesh size was employed for other part of the body.When the patient laid on the table normally, the conspicuous RF heating in the left forearm implanted with an intramedullary rod was observed at the edge of the rod (Fig. 1a). The maximum SAR was 127 W/kg which is large enough to pain the patient. When the forearm was placed on the body at the center of the scanner perpendicularly to the static magnetic field of MRI, the maximum SAR was reduced to 1.2W/kg which is less than the criteria of SAR for MR safe materials. Even when the forearm moved to right or left by 5 cm, the maximum SAR was 1.5 or 2.5 W/kg. In these positioning of the forearm at the center of the z (B0) direction, spine and abdomen can be imaged as long as the area where the artifact of the implant does not reach. To image other area, the body must be shifted. Figure 2 shows how the maximum SAR depends on the implanted forearm position along z direction: i.e. the shift of the body keeping the same positioning of the forearm. The maximum SAR increases to the position shift of 30 cm which corresponds to the edge of the transmission RF coil. Taking account into the upper limit of the local SAR at extremities (20 W/kg), tolerance of the shift is assumed ±20 cm. In the incident of the patient implanted an intramedullary rod, lumber spine imaging had been tried. In that case, our proposing positioning of the forearm can be done within the tolerance area and would have enabled the tried imaging. If this patient needs brain imaging, the body must be shifted by around 50 cm where the small SAR is shown in Fig. 2 because this position is outside of the transmission RF coil. The simulation of the normal forearm positioning at this position (50-cm shift) also showed low SAR (1.6 W/kg).The computer simulation confirmed that an intramedullary rod induces RF heating in an implanted forearm to pain the patient during an MRI examination. Positioning of the implanted forearm on the body at the center of the scanner can reduce the RF heating to the level of MR safe.