Electrocorticography grids might cause excessive heating during MR imaging

Emad Ahmadi¹, Reza Atefi¹, Emad Eskandar², Alexandra J. Golby³, Michael H. Lev¹, Rajiv Gupta¹, and Giorgio Bonmassar¹

¹Radiology, Massachusetts General Hospital, Boston, MA, United States, ²Neurosurgery, Massachusetts General Hospital, Boston, MA, United States, ³Neurosurgery, Brigham and Women's Hospital, Boston, MA, United States

Synopsis

Electrocorticography grids are routinely implanted over the cortex for pre-surgical planning in epilepsy surgery. We propose that MR imaging at 3T might cause heating injury in patients with implanted electrocorticography grids.

Purpose

To study the MR heating caused by conventional electrocorticography grids

Methods

We placed an electrocorticography grid (FG64C-SP10X-000, Ad-Tech Medical, Racine, WI) over a head phantom with MR thermodynamic properties similar to human tissues and measured the temperature increase around the grid during 30 minutes of MR imaging (Figure 1) (1). For comparative purposes we also measured temperature increases for the head phantom with no grid. A turbo-spin echo sequence was used on a 3T Siemens Trio scanner to deliver a specific absorption ratio of 4 W/kg to the phantom. Temperature changes were measured using eight optical sensors that were placed within the phantom underneath the grid. To place each sensor in its location, a tunnel was drilled inside the phantom and the sensor was passed through the tunnel. The air gap between sensors and the tunnel walls was filled with thermal conductive grease. Temperature changes during MR imaging were measured for the two cases of the phantom with and without the grid. These two experiments were done on two separate days. We kept the phantom in the MR scanner room for 48 hours before each experiment to allow the phantom temperature to reach to equilibrium with the scanner room temperature. Each temperature increase measurement was followed by a re-test measurement using the same procedures. This was done to ensure the reliability of the measurements.

Results

Table 1 shows sample temperature increases for the head phantom after 15 and 30 minutes of MR imaging for the head phantom without and with the electrocorticography grid. Table 2 shows the re-test results. Figure 2 shows the temperature increases

Conclusion

The MR heating of the head phantom with the conventional grid was more than twice as much of the phantom without the grid, which may lead to heating injury in patients with electrocorticography grids undergo MR imaging.

Acknowledgements

This research was supported by the grants 5R43NS071988-02 (NIH/NINDS), U01-NS075026 (NIH/NINDS), 1R21EB016449-01A1 (NIH/NIBIB), the National Center for Research Resources (NCRR, P41-RR14075), the National Center for Image Guided Therapy (NCIGT, P41EB015898), and by the MIND institute.

References

1. Angelone LM, Makris N, Vasios CE, Wald L, Bonmassar G. Effect of transmit array phase relationship on local Specific Absorption Rate (SAR). ISMRM Fourteenth Scientific Meeting. Seattle, USA2006.

Figures

Fig 1. The stages of making the head phantom and placing the temperature sensors inside the phantom. First, a paste was made by mixing Gracilaria Alagae powder, NaCl and water. This paste was squeezed into a mold (A) and cooled (B). The mold was then opened and the phantom was retrieved (C). Eight optical sensors were placed within the phantom (D), air gaps between sensors and phantom was filled with heat-conducting grease (E, F). Finally, the grid was placed over the temperature sensors (G).

Fig 2. MR heating of the head phantom that was scanned without grid (A) and with the electrocorticography grid (B). The middle panel shows the temperature increases sensed by the eight sensors during 30 minutes of MR imaging (D and E). The bottom panel shows the re-test temperature increases for the same three cases (G and H).

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)

2238