Heating of hip joint implants in MRI: The combined effect of RF and switched-gradient fields

Abstract

Purpose: To investigate how the simultaneous exposure to gradient and RF fields affects the temperature rise in patients with a metallic hip prosthesis during an MRI session.

Methods: In silico analysis was performed with an anatomically realistic human model with CoCrMo hip implant in 12 imaging positions. The analysis was performed at 1.5 T and 3 T, considering four clinical sequences: turbo spin-echo, EPI, gradient-echo, and true fast imaging sequence with steady precession. The exposure to gradient and RF fields was evaluated separately and superposed, by adopting an ad hoc computational algorithm. Temperature increase within the body, rather than specific absorption rate, was used as a safety metric.

Results: With the exception of gradient-echo, all investigated sequences produced temperature increases higher than 1 K after 360 seconds, at least for one body position. In general, RF-induced heating dominates the turbo spin-echo sequence, whereas gradient-induced heating prevails with EPI; the situation with fast imaging sequence with steady precession is more diversified. The RF effects are enhanced when the implant is within the RF coil, whereas the effects of gradient fields are maximized if the prosthesis is outside the imaging region. Cases for which temperature-increase thresholds were exceeded were identified, together with the corresponding amount of tissue mass involved and the exposure time needed to reach these limits.

Conclusion: The analysis confirms that risky situations may occur when a patient carrying a hip implant undergoes an MRI exam and that, in some cases, the gradient field heating may be significant. In general, exclusion criteria only based on whole-body specific absorption rate may not be sufficient to ensure patients' safety.

Keywords: MRI safety; gradient coil heating; hip prosthesis; numerical simulation; radiofrequency heating.

FIGURE 1

Human body model positions with right‐side hip prosthesis. A back view of the human body is reported. The body positions are numerated from 1 (thorax MRI) to 12 (femur/knee MRI), which correspond to the following positions of the implant head along the z‐axis with respect to the isocenter: +288 mm, +224 mm, +160 mm, +96 mm, +32 mm, −32 mm, −96 mm, −160 mm, −224 mm, −288 mm, −352 mm, and −416 mm

FIGURE 2

Maximum temperature increase after 360 seconds of exposure versus the axial position of the hip implant in the coils; z = 0 corresponds to the isocenter. Points in the plots correspond to the 12 body positions shown in Figure 1 and numerated from 1 to 12. Results for EPI, true fast imaging sequence with steady precession (TrueFISP), gradient‐echo (GRE), and turbo spin‐echo (TSE) (with constant dead time) sequences are shown, considering the effects of the RF coil alone, the GC alone, and both coils together. Symbols denote the computed values, and the interpolating lines show the trends. The four imaging regions (femur/knee, pelvis, abdomen, and thorax) related to the axial position are also indicated. Left side: Results for 1.5 T. Right side: Results for 3 T. Note that the overall maximum is plotted. This may differ only slightly from that associated with a single source, because of the different spatial distributions of temperature increase within the body associated with each source. Abbreviation: GC, gradient coil.

FIGURE 3

Plots of the temperature increase ΔT (after 360 seconds) of each voxel belonging to the region of influence versus the minimum distance d from the implant surface. Only the tissues most affected by the heating are plotted (bone, fat, muscle, and subcutaneous adipose tissue/skin). The considered ΔT thresholds of 1 K and 3 K are indicated by a black and red dashed line, respectively. Additional data for more tissues are reported in the Supporting Information. The worst conditions, in terms of maximum heating, within the four imaging regions (femur/knee, pelvis, abdomen, and thorax) are considered (see also Table 2). A, Results at 1.5 T. B, Results at 3 T. Because the appearance of the plots depends on the order in which the contributions from different tissues are added (ie, legend order: “muscle” first and “bone” last), for cases in which the same temperature/distance occurs in more than one tissue, earlier points are overwritten

FIGURE 4

Mass of tissues around the implant that exhibit a temperature increase ΔT (after 360 seconds) greater that the threshold of 1 K. From top to bottom, the group of results for 1.5 T (thorax MRI, abdomen MRI, pelvis MRI, and femur/knee MRI) and the group of results for 3 T. For the TSE sequence, only the masses computed with variable dead times for all positions are reported. Within each bar, the color indicates the fraction associated with a given tissue

FIGURE 5

Spatial evolution, at different time instants, of the temperature increase around the prosthesis for TSE, FISP, and EPI sequences (screenshots after 50 seconds, 200 seconds, and 360 seconds). The positions where the maximum temperature increase is found are considered (see Figure 2), separately, for TSE, FISP, and EPI at 1.5 T (left group of plots) and 3 T (right group of plots). For the 3T group, the results for the EPI sequence (for which GC‐induced heating prevails) are not reported, being almost identical to those at 1.5 T. The results for the TSE sequence refer to a constant dead time of 5.67 seconds