Purpose

Radiofrequency (RF) safety in MRI for medical devices like pacemakers¹ is of great importance as the worldwide population is aging and requiring more MRI scans². There is a need to assess the safety of patients carrying multiple leads^3-4, as it is not yet covered by the existing MRI compatibility standards⁵. Thus, the purpose of this study is to experimentally observe the lead coupling regarding RF-induced heating with different simplified leads, to have a better insight on the phenomenon occurring and thus approach safety guidelines suitable for every patient carrying multiple leads.

Methods

Two 40-cm long simplified leads were made of straight hardened steel wires of 1.5 mm diameter with a 0.375 mm-thick polyolefin insulation (heat-shrink tube). As these hardened steel wires which were easily available and perfectly straight were also magnetic, two other leads were made the exact same way except with non-magnetic and non-perfectly straight stainless steel wires, to check if this property affected the RF-induced heating and coupling. To verify the occurrence of a coupling phenomenon in different lead models, two other leads of the same length were made using a coaxial cable with its metallic shield removed; the stainless steel wire diameter was 0.92 mm and the PTFE insulation thickness was 1 mm. All leads were left bare along 5 mm at their extremity to represent the electrodes. In each set of leads, one lead was also left bare at the other extremity over a length of 2 mm, called the “uncapped” lead and the other, insulated, was the “capped” one, to have different termination conditions. Different configurations were studied for each set of leads (Figure 1), embedded at the right side of an ASTM gel phantom⁶ and supported by a dedicated 3D-printed plastic support and smaller 3D-printed plastic guides along the leads (Figure 2). When both leads were together, they were spaced 2.5 mm. The experiments were conducted so that there was no lead or temperature probe displacement during the measurements on the same lead (alone or in presence of the other lead). A fiber optic temperature probe (NEOPTIX T1, Neoptix, Quebec City, Canada) was positioned orthogonally on the same 3D-printed support to ensure the accuracy of measurements at the extremity of the leads. A 3’24 duration FSE-XL ASTM⁶ RF-heating sequence was then performed on the phantom inside a 1.5 T GE Signa HDx MRI scanner (General Electric Healthcare, Milwaukee, WI, USA) and the temperature was acquired in real time with the temperature probe software. The experiments were repeated, some of them several months apart, to assess the repeatability of the observations. The first leads are referred to as leads #1, the non-magnetic ones used to check the influence of this property as leads #2, and the leads made from coaxial cables as leads #3. A coupling factor CF was calculated for each lead as the ratio between the temperature elevation of the leads when they are both present and when they are alone.

Results

The temperature elevations and coupling factors are reported in Table 1. The mean value for CF_capped in the #1 leads set is 0.63±0.05 (CI_95%), and the mean value for CF_uncapped in the same leads set is 0.93±0.07 (CI_95%). Depending on the experiment, CF_uncapped values can be higher or less than 1.

Discussion

The difference in absolute temperature elevations between two measurements for the same leads can mainly arise from the position of the temperature probe at the extremity of the leads in the gel phantom. However, the experimental setup ensured a proper repeatability of the coupling factors values. The magnetic property of leads #2 does not seem to have an influence on the temperature elevations, thus leads #1 experiments are relevant to study lead coupling regarding radiofrequency. The coupling factors values on these simplified leads suggest that coupling can have a significant impact on temperature elevation, up to 40%. Moreover, the different CF_uncapped values show that lead coupling can either induce higher or lower temperatures: this highlights the complexity of this phenomenon and indicates that the presence of another lead can have an effect on RF-induced heating, and therefore should be studied carefully.

Conclusion

This study highlights the complex nature of lead coupling in MRI, and shows that the observations about lead coupling inducing higher or lower temperatures could be essential for MRI compatibility safety standards of medical devices when considering several leads. Experiments on real medical devices leads (for example pacemakers’) will be performed to verify the impact of lead coupling in more realistic leads.

Acknowledgements

The authors thank the Région Lorraine and FEDER for financial support.