MRI scanning of patients carrying implants is becoming a reality in many hospitals, because of the wide increase of population with implants, particularly in the orthopedic ones (1). The responsibility for this practice falls on the subscribing physicians or radiologists, who rely on guidance establishing safety and compatibility of implants in the MR environment (eg, (2)), which basically address issues related to magnetically induced mechanical actions and radiofrequency (RF) induced heating of tissues. Even if large-scale MR safety studies on orthopedic implants have been published that show no evidence of specific risks for patients' health (eg, (3)), additional analyses of the interaction between MR fields and metallic implants can contribute to support the scan decision on a firm scientific rationale.In a recent paper (4), we showed that the tissues surrounding a metallic hip prosthesis exposed to a gradient field (GF) may undergo a nonnegligible heating (up to some degree), as a result of the electromagnetic energy deposited in the implant. Since then, we have realized that in the MR community these results are generally received with some skepticism, probably because of the fact that most of the scientific papers dealing with thermal problems in MRI focus on RF fields (5-16). Similarly, the attention of relevant standardization and regulatory bodies is focused solely on RF-induced heating (2,17). This tendency is absolutely well-grounded when studying native tissues, where GFs are not able to develop significant thermal effects. However, in the presence of metallic foreign bodies the situation is different.Many papers show that a significant heating can be associated with the exposure of such bodies to RF fields, especially with wire-like structures, which may act as antennas (18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30). In this case, RF typically deposits a relatively low power inside the metallic parts, but their presence may cause the enhancement of the specific absorption rate directly developed in the tissues anyway. What we would like to highlight here is that, in presence of bulky metallic prostheses, GFs can produce a nonnegligible heating too, because of the thermal power generated inside the metal (Joule effect), which involves the tissues indirectly, by diffusion ("indirect effect" (31)). The GF thermal effect, which becomes sizable only for prostheses far from the coils' isocenter, has been considered up until now by a few articles (4,32-34). To sustain these findings, we exploit an analytical solution (35) that is not affected by possible artifacts of numerical results. The solution provides the current density J (in complex notation) induced within a nonmagnetic metallic sphere, with radius R, conductivity s and permeability m 0 , immersed in a nonconductive background. When the sphere is radiated by a homogeneous, timeharmonic, magnetic flux density B (peak value) at frequency f, the current density in an internal point iswhere r is the distance from the sphere center, u is the cola...