MRI Safety and Neuromodulation Systems

Introduction

Magnetic resonance imaging (MRI) has been utilized in the clinical setting for more than three decades. During this time, the technology has continued to evolve to improve image quality, acquisition time, and patient comfort. These changes have yielded MRI systems with more powerful static magnetic fields, faster and stronger gradient magnetic fields, and improved radiofrequency (RF) transmission coils. The short-term exposures to the electromagnetic fields used for MRI at the levels currently recommended in the governing MRI standard, ISO 60601-2-33, and by the United States Food and Drug Administration (FDA) have resulted in relatively few problems for the hundreds of millions of MRI examinations performed to date ( ). Most reported cases of MRI-related injuries and the few fatalities that have occurred have been due to not following safety guidelines or using inappropriate or outdated information related to the safety aspects of biomedical implants and devices ( ).

The preservation of a safe MRI environment requires constant attention to the management of patients and individuals with metallic implants and devices, because the variety and complexity of these objects constantly change ( ). To guard against adverse events and other problems in the MRI setting, it is necessary to revise the safety guidelines for implants and devices according to changes that have occurred in MRI technology and based on MRI testing best practices. In consideration of the above, this chapter discusses MRI safety issues for neuromodulation systems and presents specific information for these electronically “active” devices.

Static Magnetic Fields

The introduction of MRI technology as a clinical imaging modality in the early 1980s was responsible for a substantial increase in human exposures to strong static magnetic fields ( ). Most MR systems in use today operate with static magnetic fields ranging from 0.2-T to 3-T. The most common field strength is 1.5-T, and 3-T is the highest strength used in the clinical setting. Ultrahigh field strength MR systems currently exist in research settings, and include 4-T scanners, more than ten 7-T scanners, at least two 9.4-T scanners (University of Illinois at Chicago and Maastricht University in the Netherlands), and a 10.5-T scanner (University of Minnesota). During MRI the static magnetic field aligns the protons, typically hydrogen protons, to the direction of the magnetic field. These protons then resonate or process proportionally to the Larmor frequency. The vector sum of the magnetic moments of all of the protons is called the bulk magnetization vector.

Gradient Magnetic Fields

Gradient or time-varying magnetic fields switch rapidly in association with MRI to localize the induced signals coming from the protons in tissue, which permits spatial localization. The speed and strength of the gradient magnetic fields are limited during MRI examinations because they may stimulate nerves or muscles by inducing electrical fields in patients, resulting in peripheral nerve stimulation (PNS) ( ).

Radiofrequency Fields

The RF subsystem utilized during MRI has two phases: an excitation or transmit phase, and an acquisition or receive phase. During the transmit phase, RF energy is transmitted into the tissue of interest to tip the bulk magnetization vector out of alignment from the static magnetic field. During the receive phase, the rotating bulk magnetization vector induces current in the RF receive antenna or coil. The receive coil may be the same coil as the transmit coil (e.g., a transmit/receive (T/R) RF coil, which exists for the head, knee, wrist, and other body parts) or it may be a receive-only coil that is customized for the anatomy undergoing imaging (e.g., head, shoulder, breast, torso, etc.). In the receive-only configuration the RF coil built into the scanner, typically referred to as the body coil, is used as the transmit coil.

The majority of the RF power transmitted for MRI or spectroscopy (e.g., carbon decoupling, fast-spin echo pulse sequences, magnetization transfer contrast pulse sequences, etc.) is transformed into heat within the patient’s tissue as a result of dielectric heating ( ). Not surprisingly, the primary bioeffects and issues associated with exposure to RF radiation are related to the thermogenic qualities of this electromagnetic field.

Thermoregulatory and other physiologic changes that a human subject exhibits in response to exposure to RF radiation are dependent on the amount of energy that is absorbed ( ). The dosimetric term used to describe the absorption of RF radiation is the specific absorption rate, or SAR. The SAR is the mass normalized rate at which RF power is coupled to biologic tissue, and is typically indicated in units of watts per kilogram (W/kg). The relative amount of RF radiation that an individual encounters during an MRI examination is characterized with respect to the whole-body averaged and peak SAR levels (i.e., the SAR averaged in 1 g of tissue).

Measurements or estimates of SAR are not trivial, particularly in human subjects. Notably, the calculation is even more complicated when a metallic implant is present in a patient ( ). There are several methods of determining this parameter for the purpose of RF energy dosimetry in association with MRI procedures. The SAR that is produced during an MRI procedure is a complex function of numerous variables, including the frequency (determined by the strength of the static magnetic field of the MR system), the repetition time, the type of transmit RF coil used, the volume of tissue contained within the coil, the configuration of the anatomical region exposed, the orientation of the body to the field vectors, and other factors ( ). Each MRI system manufacturer has a different approach and makes different assumptions when calculating and reporting SAR information.

Another measurement of transmitted RF energy that device manufacturers are starting use is called B _1+RMS . B _1+RMS is the root-mean-square value of the MRI effective component of the RF magnetic (B ₁ ) field, or in other words the time-averaged RF magnetic field component relevant for creating an MR image that is generated by the MR system during a scan. B _1+RMS is measured in units of microTesla (μT). The MR scanner measures the B ₁₊ field (the positively rotating RF magnetic field produced by the scanner) needed for an imaging sequence and uses the time-averaged B ₁₊ field, or B _1+RMS , that will occur due to a particular imaging sequence. Thus, the B _1+RMS value is calibrated by the MR system’s software during the “prep” or “prescan” phase or measurements of an MRI exam. An important characteristic of B _1+RMS is that it is not an estimated value but a known quantity based on the pulse sequence and associated parameters. Furthermore, B _1+RMS is not patient dependent, nor is it calculated differently by a given MR system manufacturer. Because B _1+RMS is considered to be a more precise RF exposure metric than SAR, device manufacturers have begun to use values for B _1+RMS that must not be exceeded when scanning patients with active implants. When following B _1+RMS as a condition of use, the SAR value is irrelevant, unless of course a particular operating mode is specified. Importantly, using B _1+RMS tends to provide better “performance” for the MRI examination because it has fewer limitations versus using SAR values for patients with implants and devices.

MRI Safety and Screening Patients for MRI Procedures

The establishment of thorough and effective screening procedures for patients and other individuals is one of the most critical components of a program that guards the safety of all those preparing to undergo MRI procedures or enter the MRI environment ( ). An important aspect of protecting patients and individuals from MR system-related accidents and injuries is understanding the risks associated with the various implants, devices, accessories, and other objects that may cause problems in this setting. This requires obtaining accurate information and documentation about these objects to ensure the safest MRI situation possible. In addition, because MRI-related incidents have been due to deficiencies in screening methods and/or a lack of properly controlled access to the MRI environment (especially with regard to preventing personal items and other potentially problematic objects going into the scanner room), it is crucial to set up procedures and guidelines to prevent such incidents from occurring. Importantly, many guidelines and recommendations have been developed to facilitate the screening process (see www.MRIsafety.com ) ( ).

Interactions Between the MRI Environment and Implanted Medical Devices

Injuries or Device Damage Related to the Static Magnetic Field

Magnetic-field-related issues are known to present hazards to patients with certain implants or devices, primarily due to movement or dislodgment of objects made from ferromagnetic materials ( ). Medical devices with ferromagnetic materials will be attracted to and aligned with the strong static magnetic field of the MR system ( Fig. 24.1 ). For implants, this movement could result in patient injury (e.g., torn sutures, internal bleeding, etc.) or the need for surgical revision to reposition the device to restore therapy or for patient comfort. Exposure to the power static magnetic field may also alter the device settings for an electronically activated device, requiring reprogramming, or permanently damage the device, necessitating replacement surgery ( ).

Numerous studies have assessed magnetic field interactions (i.e., translational attraction and torque) for implants and devices by measuring translational attraction and torque associated with the static magnetic fields of MR systems. These investigations demonstrated that for certain items MRI examinations may be performed safely if the items are nonferromagnetic or “weakly” ferromagnetic (i.e., the object minimally interacts with the magnetic field in relation to its in vivo application), such that the associated magnetic field interactions are insufficient to move or dislodge the items in situ. Furthermore, the “intended in vivo use” of the implant or device must be taken into consideration, because this can affect whether or not a particular item is acceptable in a patient undergoing an MRI examination. Notably, sufficient counterforces may exist to retain even a ferromagnetic implant in situ, such as those associated with fibrous encapsulation of a pulse generator used with a neuromodulation system.

Injuries or Device Damage Related to Gradient Magnetic Fields

The MRI environment may be unsafe for patients with certain biomedical implants or devices due to the induced energy and/or eddy currents related to the fast-switching gradient magnetic fields that are used for spatial localization as part of the imaging process. Gradient magnetic fields used during MRI switch or “slew” between positive and negative polarity. According to Faraday’s Law of Induction, any circuit within these rapidly changing magnetic fields is subject to induced currents. The fluctuating magnetic field can be generated with a fixed magnet and moving the circuit through the magnetic field, as is done by a typical electric generator configuration, or by using a stationary circuit and varying the magnetic field, as is the case with an MRI examination.

The switching gradient magnetic field can induce a voltage along conductive structures that may be located within an MR system ( Fig. 24.2 ). With typical neuromodulation systems that utilize a lead and an electrode, the circuit can be formed between the tip of an implanted electrode and the pulse generator, on metallic structures, or on individual components (specifically coiled structures like antennae or energy transfer coils). If the voltages are high enough or the resistance is low enough, current will flow. According to Faraday’s Law, the larger the loop area the greater the induced voltage. A variety of interactions can occur if current flows. For electronically activated devices like cardiac pacemakers and neurostimulation systems, there is the risk of unintentional stimulation at the switching rate of the gradient coils.

Unintentional stimulation can be relatively benign, as is the case with most neurostimulation scenarios, since neurostimulation pulse frequencies are much higher than the gradient switching frequency associated with an MRI examination. In other cases, unintentional stimulation can pose a significant health risk for the patient, such as with cardiac pacemakers. It should be noted that the gradient-induced voltages acting on the antennae and recharge coils of electronically activated medical devices such as cardiac pacemakers and neurostimulation systems can be quite high, due to the number of coils within the associated components. If the circuitry is not robust enough to shield against these high voltages, the medical device may sustain damage to its communication or recharge system, memory, or output channels in association with an MRI examination ( Fig. 24.3 ). If the device is damaged, it may require reprogramming or surgical replacement.

Excessive Heating and Device Damage Related to RF Energy

The use of transmit RF coils, physiologic monitors, electronically activated devices, and external accessories or objects made from conductive materials has caused excessive heating, resulting in serious burn injuries to patients undergoing MRI procedures ( ). The heating of implants and devices may also occur in association with MRI, but this tends to be problematic primarily for objects made from conductive materials that have an elongated shape, such as electrodes, leads, guide wires, and certain types of catheters (e.g., thermodilution catheters with thermistors or other conducting components), or those forming a loop of a certain diameter ( ).

In general, published reports have indicated that only minor temperature changes occur in association with MRI examinations involving relatively small metallic objects that are “passive” implants (i.e., those that are not electronically activated), including implants such as aneurysm clips, hemostatic clips, prosthetic heart valves, vascular access ports, and similar devices. Thus, the heat generated during MRI involving a patient with a “small” passive metallic implant does not represent a substantial hazard. In fact, to date there has been no report of a patient being seriously injured as a result of excessive heating that developed in a relatively small passive metallic implant or device.

However, long conductive structures typically act as more efficient unintentional antennae. These unintentional antennae collect the RF energy associated with MRI and dissipate it at the path of least resistance. For leads used with neuromodulation systems this typically occurs at the electrodes of the leads, which are commonly found at the distal end and in contact with the organ area receiving stimulation (e.g., brain, occipital nerve, vagus nerve, sacral nerve, epidural space in the spine, etc.). Since the electrodes used with neuromodulation systems are relatively small, the dissipated energy is concentrated very close to their surface ( Fig. 24.4 ). Therefore, significant heating can be generated under certain clinical conditions and there have been several reports of serious patient injuries ( ). A variety of variables contribute to the potential heating of a neuromodulation system related to the use of MRI ( ). The critical variables that impact MRI-related heating of an implanted medical device are presented in Table 24.1 .

Table 24.1

Critical Variables That Impact MRI-Related Heating of a Medical Implant

Area	Variable	Comments
MRI-related variables	RF frequency	The RF frequency is related to the static magnetic field strength of the MR system. At 1.5-T 64 MHz is used and at 3-T 128 MHz is used for hydrogen proton imaging.
	RF power level, whole-body averaged specific absorption rate (SAR) or B 1+RMS	In the normal operating mode for the MR system, the RF power is limited to 2.0 W/kg whole-body averaged SAR and 3.2 W/kg average head SAR. In the first-level controlled operating mode, the RF power is limited to 4.0 W/kg whole-body averaged SAR and 3.2 W/kg average head SAR. See discussion regarding SAR variability concerns and preference for the use of B 1+RMS .
	Landmark (area of body exposed to RF energy)	Depending on which area of the body is being imaged, more or less of the implanted medical device may be exposed to the RF energy.
	Transmit RF coil size	Transmitting RF with an extremity or head exposes a lesser extent of the patient’s body, and therefore potentially less or none of the implanted medical device.
	Scan duration	Thermal injuries are based on thermal doses to tissue. Thermal dose is a function of time and temperature, so it takes a shorter time to damage tissue at higher temperatures.
Medical device variables	Conductive structure design	Some design structures may be better or worse antennae.
	Conductive structure length	Structures may have a worst-case length depending on the resonant frequency. Short structures may not be long enough to act as a good antenna.
	Implant location	The amount of RF energy incident on a device depends on the device location, including lead routings.
	Loops/coils of the leads	Cross-over points of the leads provided by strain relief or “slack” loops can alter the distribution of RF energy, cause some localized RF energy dissipation at the cross-over point, and decrease the quantity dissipated at the electrodes.
Patient variables	Body size	Larger patients have more tissue, and therefore more RF power is needed to excite the tissue for MRI.
	Tissue composition	Fat is more permeable to RF energy, so less energy is required to penetrate it compared to a muscular patient of the same size.

All these variables must be taken into consideration to create safe operating conditions for patients with neuromodulation systems undergoing MRI. It is important to appreciate the complex interactions between the variables related to RF heating. In fact, each RF frequency (i.e., 64 MHz associated with 1.5-T, 128 MHz associated with 3-T, etc.) needs to be tested individually for a given neuromodulation system, because the resulting temperature rises may actually be less at higher frequencies depending on the length of the lead or other conductor ( Fig. 24.5 ) ( ).

Different mitigation strategies have been employed to create neuromodulation systems that can be labeled MR Conditional. For example, some devices have MR Conditional labeling that excludes the RF exposure entirely by limiting the MRI examination to the use of a head and/or extremity T/R RF coil (i.e., if the neuromodulation system is implanted outside that particular T/R RF coil). Others use a reduced RF-transmitted power limit to minimize exposure to the RF energy, while still other neuromodulation products have been specially designed to manage the RF energy safely so the MR system can be operated under typical RF power levels (normal operating mode or first-level controlled operating mode).

Additionally, the use of the whole-body averaged SAR value reported by the MR system itself is especially problematic with regard to MRI-related implant heating ( . The MR system’s reported SAR values can be considered to be relatively conservative, “not-to-exceed” values, such that each MR system manufacturer may have vastly different safety margins to ensure it does not exceed the displayed SAR value(s). Therefore, it is important to understand that implant heating may differ significantly when using different MR systems of the same static magnetic field strength and frequency. As previously indicated, the variation is due to the different methods and assumptions that MR systems use to estimate and control SAR. Significant variation in implant heating for a deep brain stimulation (DBS) lead in association with different 1.5-T scanners (notably from the same manufacturer) was first reported by , and further examined in other investigations ( ). The specific concern is when MRI safety labeling is based on data generated using a particular MR system with a large SAR safety margin, which equates to lower transmitted RF power. Subsequently, in the clinical setting, the device’s safety conditions may be complied with on an MR system with a much lower SAR safety margin, equating to a higher RF power level for the same MR system displayed SAR. The result could be a much higher temperature rise for the medical implant.

In addition to the concern about RF heating of medical devices, the RF energy collected by the creation of unintentional antennae can travel in the opposite direction toward the pulse generator of a neurostimulation system. This aspect RF energy can damage device electronic components or memory, and may necessitate device replacement surgery in the event that the electronics are compromised by high-frequency RF energy ( ).

Evaluation of MRI Issues for Neuromodulation Systems

The evaluation of various MRI issues that affect a neuromodulation system is not a trivial matter, and in fact may be quite challenging ( ). The proper assessment of a medical product typically entails characterization of magnetic field interactions (translational attraction and torque), MRI-related heating, induced electrical currents, and artifacts ( ). Furthermore, a thorough evaluation of the impact of MRI conditions on functional and operational aspects is also necessary. Importantly, a neuromodulation system demonstrated to be acceptable for a patient according to one set of MRI conditions may be unsafe under more “extreme” or different conditions (e.g., higher or lower static magnetic field, higher or lower RF wavelength, greater level of RF power deposition, faster gradient magnetic fields, use of a different transmit RF coil, etc.). Accordingly, the specific test conditions that are utilized for a given neuromodulation system must be based on carefully conducted test procedures before making a decision regarding whether it is safe for a patient to undergo an MRI examination.

Historically medical implants, including many neuromodulation systems, were tested according to documents developed by the American Society for Testing and Materials International ( ). As MR systems evolve, the medical device industry has come to a better understanding of the possible interactions between implantable medical devices and the issues associated with MRI, which in the case of neuromodulation systems are particularly complex. Consequently, a group consisting of representatives from medical device and MR system manufacturers, MRI scientists and engineers, and regulatory agency representatives developed a test specification specifically related to evaluating the safety of “active” implantable medical devices in association with the use of MRI. This work took a number of years to complete. The first edition of the document, ISO/TS 10974, was published in May 2012. This specification is now recognized by most regulatory agencies around the world as the preferred test methodology to generate the safety evidence necessary for MR Conditional approval with respect to active medical devices.

Newly developed implants and devices, as well as devices already on the market, are being tested for MRI issues on an ongoing basis. This necessitates continuous endeavors to obtain current documentation for these devices prior to subjecting a patient to an MRI examination. In addition, the nuances of MRI testing, especially with respect to the evaluation of MRI-related heating and the identification of alteration in function for active devices, and the terminology applied to label these medical products must be understood to facilitate patient management ( ). Importantly for neuromodulation systems, the labeling that ensures acceptable use of MRI always presents many specific conditions to ensure patient safety. Any deviation from the defined procedures can lead to deleterious effects, severe patient injuries, or fatalities ( ). As previously stated, the preferred source for obtaining the most current MRI labeling information is the medical device manufacturers. However, for convenience, when information is available it is published to healthcare professionals and others in a reference manual ( ) and online at www.MRIsafety.com .

MRI Procedures and Neuromodulation Systems

Neuromodulation systems have been employed for a variety of neurological disorders and other conditions. A typical neurostimulation system consists of an implantable pulse generator (IPG; similar to a cardiac pacemaker) and a set of wires or leads that conduct the electrical pulses to the therapeutic target via one or more electrodes. The IPG typically contains electronics and a single-use or rechargeable battery. The size of the IPG and the therapeutic target dictate the potential implant locations and lead routings through the body. Current therapies approved by the FDA include spinal cord stimulation (SCS) for chronic pain; DBS for essential tremor, Parkinsonian tremor, dystonia, and obsessive–compulsive disorder; vagus nerve stimulation (VNS) for epilepsy and depression; sacral nerve stimulation for urinary and fecal incontinence; and gastric stimulation for gastroparesis.

During the early days of MRI, because of the inherent risks associated with neuromodulation systems in the MRI environment, the presence of these electronically activated implants and others was considered a strict contraindication for patients. However, over the years various studies have been performed that indicated relative safety for neuromodulation systems ( ). In fact, many neuromodulation systems have received FDA approval for MR Conditional labeling for MRI. As such, if the specific guidelines are followed, MRI examinations may be conducted safely in patients with these neuromodulation systems.

A number of tactics can be employed to allow safe MRI examinations to be performed in patients with neuromodulation systems. Some limit the strength of the static magnetic field, gradient magnetic fields, and/or the RF fields at certain levels, and may also limit the landmarks or portion of the body that can be imaged. Various representative examples of neuromodulation systems that have criteria defined to permit safe MRI examinations are presented in this chapter, with the acknowledgment that many other devices exist with approved MR Conditional labeling claims. When available, labeling approved by the FDA is presented. It should be noted, however, that certain neuromodulation systems have approval outside the United States for patients referred for MRI examinations. Importantly, in certain cases a particular neuromodulation system may be labeled MR Unsafe in the United States but MR Conditional outside the United States. Therefore, it is vital to obtain country-specific labeling information for a given neuromodulation system. As previously noted, healthcare professionals are advised to contact the device manufacturer to obtain the latest information to ensure patient safety relative to the use of MRI.