4178

Mitigating heating risk for long active metallic-backbone or metallic-braided cardiovascular devices using miniaturized Baluns (MBaluns): Design parameters, heating tests, and swine validation
Akbar Alipour1, Eric S Meyer1, Wolfgang Loew2, Ronald D Watkins3, Hassan Elahi1, Aravindan Kolandaivelu1, Susumo Tao1, Michael A Guttman1, Jeff Schweitzer4, Gregory Olson4, Henry R Halperin1, and Ehud J Schmidt1

1School of Medicine, Johns Hopkins University, Baltimore, MD, United States, 2Cincinatti Children’s Hospital Medical Center, Cincinnati, OH, United States, 3Stanford University, Stanford, CA, United States, 4Abbott Laboratories, Minnetonka, MN, United States

Synopsis

Long (>wavelength/4) actively-tracked metallic-backbone or metallic-braided cardiovascular devices are not used in MRI-guided interventions due to surrounding-tissue heating concerns. Such devices may be used if induced currents on the metallic-backbone and internal cables are sufficiently attenuated. At ISMRM 2018 we demonstrated a miniaturized resonant floating Balun (MBalun), and two actively-tracked MRI-guided metallic interventional devices built using MBaluns. MBaluns were constructed with loosely-wound solenoids overlaid on the metallic-backbones, generating strong transverse magnetic fields that attenuated surface currents and internal-cable common-mode currents. We now provide electromagnetic simulation revealing MBalun critical dimensions, along with phantom heating and swine experiments that validate the designs.

Introduction

Cardiovascular catheters and guidewires, long (>wavelength/4) and narrow structures, are preferably constructed of metallic alloys. The MRI’s transmission-coil induces currents on the metal surface and common-mode (CM) currents in the wires within, which may heat surrounding tissues.1 Therefore, these devices are not currently available for use in MRI. Several solutions are used to avoid the induced currents in MRI-compatible catheters. Current solutions are; (a) braids from non-metallic materials,2 (b) metallic-backbone devices physically bisected at wavelength/4 increments,3 (c) metallic-backbone devices restricted for use with low SAR sequences.4 Solution (a) requires significant space within the device for heat-mitigation on each cable, and may reduce the fidelity of desired differential-mode signals if transmission losses are associated with the heat-mitigation method. Floating resonant RF traps (Baluns) are currently employed to reduce current induction into the shielded multi-core cables that carry information from surface-array coils to the scanner’s receivers.5 Miniaturizing conventional Balun size might form a solution for metallic-bone devices, but a recent study found such a reduced-size Balun provides only ~3 dB attenuation.6 At ISMRM 2018 we demonstrated a new Miniature Balun (MBalun) that enabled use of elongated actively-tracked metallic-backbone interventional devices.7 We employ electromagnetic simulations to reveal key MBalun design parameters, in order to construct devices dimensionally similar to their non-MRI-compatible analogs.

Method

The MBalun is constructed of two concentric metal layers with an insulator layer between them. The inner layer is a metal tube or a metallic braid, while the outer layer is a multi-turn spiral-wound copper wire (Fig.1A). The volume-integrated transverse magnetic field (MTr) in the MBalun produces a magnetic moment (Dm): $$D_{m}=\int_{}^{} M_{Tr}dV_{MB}, V_{MB}=\pi(r_o^2-r_i^2) L_{MB}$$ where LMB and VMB are the MBalun length and magnetically-effective volume and ro, ri are its outer and inner radii, respectively. A wire carrying a current I that passes through the MBalun couples a flux φ. Therefore, considering an MBalun surrounding a current-carrying wire, the effective inductively-coupled inductance generated by the MTr is: $$L_{e}=\frac{\phi}{I}=\frac{N\mu D_{m}}{I L_{MB}}$$ where N is the winding number and μ the permeability. As a result, if Dm is evaluated in electromagnetic simulation, can be computed. The impedance of the resonant MBalun at its resonant frequency (wRes) is given by: $$Z= \frac{(w_{Res}^2 L_e^2)}{R}$$ where $$$R\sim\frac{N}{\delta_{c} \sigma_{c} 2\pi r_{o}}$$$ is the MBalun’s resistance, $$$\delta_{c}$$$ the skin depth, and $$$\sigma_{c}$$$ the wire conductivity.

MBalun performance relies mainly on its effective inductance (Le), which depends on three design parameters: the normalized insulator thickness ($$$\triangle=(\frac{insulator-thickness}{inner-tube-diameter})$$$), number of windings (N), and the normalized pitch ($$$p=(\frac{Dist}{Diam})$$$), where Dist and Diam are the wire diameter and the distance between successive windings. The current running inside a tightly-wound (p≈1) solenoid produces a strong magnetic field along the solenoid axis (Mz) and almost no MTr (Fig.1B), which therefore does not couple efficiently to the magnetic-fields produced by current-carrying wires running inside. Current in a leaky (p>1) solenoid generates strong MTr (Fig.1C), which allows the solenoid to effectively couple to magnetic-fields produced by wires running inside. A series of electromagnetic simulations (CST, Germany) were carried out to investigate the effects of design parameters N, ∆, and p on the properties of a single MBalun. MBaluns were designed to provide possible solutions for (a) a metallic-tube active guidewire (Fig. 2A) and (b) a metallic-braided electrophysiology (EP) catheter (Fig. 2B). SAR simulations (CST) were conducted to evaluate current suppression on a wire passing through an MBalun. Heating tests were performed on both devices in a gel phantom during high SAR sequences in a 1.5T MRI. The devices were evaluated for ease of cardiovascular navigation, imaging conspicuity, tracking precision, and heating in swine models using breath-held ECG-gated SSFP sequences.

Results

The effect of MBalun parameters was numerically analyzed for different N, ∆, and p values. The effective-inductance Le decreased as ∆ increased (Fig. 3A). Le was maximal at p=3 (Fig. 3B). The SAR around a 35-cm wire mounted with three Mbaluns (Fig. 3C) was reduced by ~30.3 dB. In ASTM gel-phantom heating tests, both devices incurred temperature increases (Fig. 4A, B) within ASTM/FDA/IEC limits. Both the guidewire (Fig. 5A) and the EP catheter (Fig. 5B), permitted rapid vascular navigation resulting from good MR-Tracking visibility and improved torquability (Fig. 5C, D). The device shafts appeared dark, indicating that surface currents on the metallic backbones and metallic braids were effectively suppressed.

Conclusion

A new method for ameliorating the risk of heating for long metallic-backbone devices was devised and validated. MBaluns provide for efficient heat-mitigation with an ~20% cost in device dimension.

Acknowledgements

This study was supported by NIH R01-HL094610 and Abbott Laboratories grants.

References

[1] Nitz WR, Oppelt A, Renz W, Manke C, Lenhart M, Link J. On the heating of linear conductive structures as guide wires and catheters in interventional MRI. J Magn Reson Imaging. 2001;13(1):105-114.

[2] Kochli VD, McKinnon GC, Hofmann E, von Schulthess GK. Vascular interventions guided by ultrafast MR imaging: evaluation of different materials. Magnetic resonance in medicine 1994;31(3):309-314.

[3] Basar B, Rogers T, Ratnayaka K, Campbell-Washburn AE, Mazal JR, Schenke WH, Sonmez M, Faranesh AZ, Lederman RJ, Kocaturk O. Segmented nitinol guidewires with stiffness-matched connectors for cardiovascular magnetic resonance catheterization: preserved mechanical performance and freedom from heating. J Cardiovasc Magn Reson. 2015;17:105.

[4] Campbell-Washburn AE, Rogers T, Mazal JR, Hansen MS, Lederman RJ, Faranesh AZ. Real-time inversion recovery for infarct visualization during MR-guided interventions. J Cardiovasc Magn Reson. 2016;18(Suppl 1):P205.

[5] Seeber DA, Jevtic J, Menon A. Floating shield current suppression trap. Concepts in Magnetic Resonance Part B: Magnetic Resonance Engineering 2004;21B(1):26-31.

[6] Griffin GH, Anderson KJ, Wright GA. Miniaturizing Floating Traps to Increase RF Safety of Magnetic-Resonance-Guided Percutaneous Procedures. IEEE Trans Biomed Eng. 2017;64(2):329-340.

[7] Schmidt EJ, Meyer ES, Watkins RD, Elahi H, Loew W, Schweitzer J, Olson G, Kolandaivelu A, Halperin HR. Actively-tracked metallic electrophysiology catheters and guidewires with miniature floating radio-frequency traps: Theory, Design and Validation, Proceedings ISMRM 2018.

Figures

Figure 1: (A) Three-dimensional drawing of an MBalun, illustrating design attributes. The two layers are shorted at one end and resonated with a tuning capacitor at the other end. (B) The magnetic field produced by current flowing in a tightly wound, p≈1 solenoid is mostly along the solenoid axis (Mz), with negligible MTr. This solenoid cannot effectively couple with the magnetic field produced by a current-carrying wire running inside (black arrow), which is mostly MTr. (C) For a loosely-wound (p>1) solenoid, a significant MTr is present, so it can couple effectively with the magnetic field produced by the current-carrying wire.

Figure 2: (A) Expanded MBalun including a solenoid and a 9-pF thin-film resonating capacitor used in a small-diameter metallic-tube active guidewire instrumented with two distal MR-Tracking micro-coils. (B) Expanded MBalun including 36-pF thin-film resonating capacitor and a solenoid used in a metallic-braided deflectable EP ablation catheter with four distal MR-tracking micro-coils.

Figure 3: (A) Effect of insulator thickness (∆=0.2-0.6). With other MBalun parameters (N\p=125\3) kept constant, the total inductance increases with ∆, since the MBalun’s outer-diameter grows. On the other hand, the effective-inductance Le decreases as ∆ increases since windings are at a large angle relative to the shaft at low ∆. (B) Effect of normalized pitch (p) with p=1-5, with N and ∆ kept constant (N\∆=125\0.4). Le decreases above p=3 due to increasing field inhomogeneity. (C) Effect of three MBaluns on the SAR outside a current-driven wire. The initially-strong SAR, is attenuated by ~30.3 dB immediately after the third MBalun.

Figure 4: Gel-phantom temperature test results using catheters placed far from the magnetic isocenter along the Left-Right and Superior-Inferior directions, where body-coil induced electric fields were at their maximum. Fiber-optic temperature sensors were placed at four locations close to the device tips, during 15 min of high SAR (3.99 W/Kg) imaging in an ASTM gel phantom. (A) MBalun-equipped guidewire heating, which was ~1°C at the tip. (B) MBalun-equipped EP catheter heating, which was ~0.7°C at the tip.

Figure 5: (A) A 1.1 mm outer diameter (OD) active guidewire with 8 MBaluns on its shaft (M1-M8) mounted on a nitinol tube. The tube includes two 44-AWG cables that carry the MR-Tracking signals. (B) 2.6 mm OD deflectable active EP catheter with 11 MBaluns on its shaft (M1-M11). (C), (D) Breath-held ECG-gated SSFP (TR\TE\FA=3.5ms\1.5ms\45°, matrix=128x256, FOV=23x30 cm2) of the active guidewire (C) and the active EP catheter (D), combining all receiver-channels (spine-array, body-array, 4 MR-Tracking coils). Yellow arrows show locations of tracking coils, and red arrows show absence of "over-flipping".

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)
4178