Safe MRI in Patients With Cardiac Devices: It Takes a Team

Robert Watson, MD, PhD; Paul Friedman, MD; Suraj Kapa, MD; Nancy Acker, RN; Connie M Dalzell, RN


August 25, 2016

Editorial Collaboration

Medscape &

This feature requires the newest version of Flash. You can download it here.

Suraj Kapa, MD: Greetings, my name is Suraj Kapa, and I'm a cardiologist at Mayo Clinic. During today's Round Table Review, we will be discussing [magnetic resonance imaging] MRI in patients with devices. I am joined by my colleagues, Dr Paul Friedman, vice chair of cardiovascular diseases; Dr Robert Watson, chair of neuroradiology; and registered nurses Connie Dalzell and Nancy Acker, who specialize in pacing. Welcome.

All: Thank you.

Dr Kapa: As a general question, why are you worried about MRI in patients with devices?

Paul A. Friedman, MD: Historically, patients were scanned, at times, without awareness that they had implantable devices. Those devices weren't designed to tolerate MRI scans. The scans can generate currents in the devices, and those currents can induce arrhythmias. They can cause cautery to the heart muscle, and there were cases of fatalities because dangerous arrhythmias were induced in patients.

It's important to note that that was in the 1980s.

Since then, devices have undergone significant evolution. Since at least the early '90s, with new pass-through filters and other technologies—although they vary by the manufacturer—have been much more resistant to external electromagnetic interference.

Defining MRI-Safe, MRI-Unsafe, MRI-Conditional

Robert Watson, MD: That's correct. A number of those devices, it's important to note, while they have improved in shielding, prior to 2011 were truly still regarded as "MRI-unsafe." Only in 2011 did we get the new influx devices that are considered to be "MRI-conditional."

Dr Friedman: It's worth emphasizing that. We could broadly categorize the devices into two categories: "MRI-conditional," specifically engineered to be in an MRI-scanner tested and approved for scanning under specific conditions, and what I'll refer to as legacy devices (although they can be new devices) that were not designed to be scanned. As we'll discuss, we've scanned close to 1000 such patients, but in fact, it requires special precautions and considerations.

Maybe it would be good for you to review [the terms] "MRI-safe," "MRI-unsafe," "MRI-conditional." What do those terms mean to a radiologist?

Dr Watson: I'm happy to do that. There are many implanted devices that are truly "MRI unsafe"; under no conditions should someone go into an MRI scanner with those devices. Ferromagnetic oxygen tanks and such in an MRI environment are inherently unsafe.

There are other devices that are "MRI-safe," such as those made of plastic, in MRI. They have no risk associated with them.

Implanted devices that have any metal in them, under certain conditions, can be scanned safely. Broadly, these types of devices, like vagal-nerve stimulators and, as you say, the newer generation of pacemakers, are considered "MRI-conditional," which means, under the appropriate conditions, we can scan them safely.

One term we really try to react against is "MRI-compatible" because that's a squishy term. It doesn't tell you whether it's safe or unsafe or the conditions that we need to scan it safely with.

Nancy Acker, RN: One important point to add is that when you have an "MRI-conditional" device, the whole component of the device, the leads included, have to be considered "MRI-compatible." If you have a device that is itself MRI-conditional but the leads are not, then it is no longer considered "MRI-compatible."

Dr Watson: That's correct. The entire system has to be part and parcel of the conditional system; that's correct.

Dr Kapa: Can you give us an idea of the Mayo Clinic experience and approach to patients coming with a device—whether conditional or presumptively safe—who need an MRI?

Dr Friedman: Sure. A clinician would order an MRI scan for a patient because there's a clinical need, and then a colleague in radiology would determine whether there's an alternative imaging modality that can be undertaken. If the answer is no, then we have a specific protocol in place that involves radiology, radiology physics, and electrophysiology. Essentially, the patient goes through a checklist—a formalized process that we periodically review. By doing that, we've safely scanned almost 1000 patients.

The first step is to assess the device, and I'll ask Connie to comment. When a patient is sent for an MRI scan, what are the steps you go through? What do you look at in the device? What requires additional action? How do you program it, and what kind of monitoring do you do during the scan?

Getting Patients Ready for the Scan

Connie M Dalzell, RN: Sure. When the patients present to Mayo, the pacemaker nurses evaluate the device. We're looking [to rule out] pacemaker dependence—if they have a heart rate of 30 beats a minute or faster, they're not pacemaker-dependent. We also look at the device itself and how it's functioning, whether the battery is at elective replacement or has enough battery reserve left. We also test the leads to determine the sensing and capture thresholds to make sure that the device is functioning appropriately before the patient even gets to the MRI suite.

Dr Friedman: A few key points: the patient has to have their own heartbeat, meaning not pacemaker-dependent. If they are pacemaker-dependent, then if they have an "MRI-conditional" device, one that's approved, we scan them anyway. If they are dependent and they don't have a conditional device—so it's a standard pacemaker defibrillator—next step is what?

Ms Dalzell: The next step on a patient who is pacemaker-dependent but doesn't have an "MRI-conditional" device is to set them up with one of our heart-rhythm services physicians to review the patient's history, the need for the MRI, and then talk with the patient about the risks associated with an MRI if, in fact, we do go forward with the scan.

Dr Friedman: That's really been a tiny number, in our experience. Then during the scan, Nancy, what kind of monitoring are you doing? What actual physiologic signs are you monitoring? What are you looking for?

Ms Acker: The patient is hooked up to an ECG monitoring system that can go into the MRI suite. We are on the outside of the MRI suite, monitoring their heart rate and rhythm, and then also they have a pulse-oxygenation device on that monitors their pulse oxygenation during that time period. We watch throughout the whole entire scan. We're looking for any arrhythmias. We're looking for any abnormalities in device function. Then, when the patient is removed from the scan, then we again check the device to make sure all the [parameters] we checked preprocedure look good again postprocedure.

Dr Watson: Right. I would add to that, while the patient is being scanned, they are holding a squeeze ball. If they feel anything unusual, any discomfort, or anything that they have any question about, then they can squeeze that ball, which sets off a loud alarm at the control panel. The technologists are under instruction to immediately stop the scan, make contact with the patient and, if need be, get the patient out of the room.

Ms. Acker: Correct.

Diverting RF Energy and Working With Magnetic Fields

Dr Friedman: Now before the scan, the radiology team has a physicist involved to consider the use of transmit-receive (TR) coils. Tell us [about the role of the] physicist and these TR coils and why they are important.

Dr Watson: We are blessed with a great team of MRI physicists. Our protocol at Mayo—one thing that we think is important and sets our program apart in establishing a level of safety—works to keep our deposited [radiofrequency] RF energy <1.5 W/kg SAR or specific absorption rate during the scan.

You can also be smart about the coils that you use, in trying to keep the RF energy away from the device as much as possible. The more [scans we do, the more we learn] that we don't always have to scrupulously use a transmit-receive head coil, for example, which we did in the early stages to keep all the RF away from a device in the chest. We started when we did head exams, just limiting our practice to using transmit-receive head coils.

Now we have branched from that where we will use what's called "body transmit," where the MRI is depositing some RF energy over the device. But we keep that <1.5 W/kg, because, not only could the RF affect the device itself, but there are [also] concerns that as RF energy is deposited on leads, it can set up heating potential in the leads. We want to maintain a large cushion for safety, and that's what our physicists are doing. As radiologists, when we protocol an exam and we need T2-weighted sequences, [fluid attenuation inversion recovery] FLAIR, and diffusion, then the physicist is working hand in glove with us to keep the parameters of the scan and the RF energy at a level that we are comfortable with.

Dr Friedman: Now you've mentioned RF energy, and that's the one that we worry about the most with devices. But there are three kinds of fields, so tell us how MRI scanners are rated. For example, 1.5 Tesla—what does that mean, and what are the other kinds of fields they generate?

Dr Watson: Right. The Tesla is an indication of the static magnetic field. In clinical use, high-field-strength magnets that are used most frequently are 1.5 Tesla and 3 Tesla. We've been, heretofore, limiting our practice to 1.5 Tesla scanners. That's an indication of the main magnetic field in the room.

The other thing to consider is a time-varying gradient magnetic field. Within the MRI, there are so-called "gradients," which change at a very rapid rate. The take-home message is what they serve to do: allow spatial encoding of the signal, such that you know where in 3-dimensional space a certain soft-tissue pixel, for example, is in your image.

[In summary,] there's a complex interplay of the main magnetic field, the time-varying gradients, and the deposited RF energy—all of which could impact an implanted cardiac device.

Dr Friedman: But in case of emergency, two of those can be turned off [quickly] and one can't, correct?

Dr Watson: Exactly. The time-varying gradients and the RF can be turned off, but the main magnetic field is always on; MR safety 101, the magnet is always on.

Dealing With Potential Complications

Dr Friedman: Tell us have we ever had any emergencies or any situations that come to mind from a nursing standpoint? How often have we had to stop a field? I'm not aware of any medically important complications from scanning with the screening we've put in place.

Ms Dalzell: I can think of devices that have had what we call "power-on reset." Those are [older] devices—from 2005, 2006, and earlier—that are electrically set into a backup mode. We have pulled patients out of scanners when that happens, checked the devices to make sure that they are functioning [properly], and then proceeded with the scans. Do you know of any others, Nancy?

Ms Acker: The only other thing I can think of is that, patients might feel a little uncomfortable. A lot of the patients are put into an asynchronous mode, meaning that the pacemaker doesn't [monitor] what their heart is doing. It just paces at a set mode and at a set rate. Some patients become anxious in their procedure, and their sinus rate could rise above the programmed rate, and sometimes they can get what we call "pacemaker syndrome" from that. Patients who have experienced that have communicated to us that they haven't felt good by squeezing the ball. Programming that rate higher has made them feel better. Other than that, I can't think of any adverse effects that patients feel from this device.

Dr Kapa: Do you see any differences—on the programming side or otherwise—between pacemakers, resynchronization devices, and defibrillators?

Ms. Dalzell: When we scan patients with defibrillators, there is one major difference, which is that it can deliver a shock if it sees a tachyarrhythmia. So, with those devices, we turn tachycardia detection off while patients are in the scan. Otherwise, as far as pacing, we follow a protocol for how that gets programmed, [depending on] whether it's a pacemaker or a defibrillator, whether it's single-chamber, dual-chamber, or a biventricular device.

Dr Kapa: How about recently implanted devices, if someone just had a device implanted, say, a week ago?

Dr Friedman: We've done approximately 20 where, for medical urgency, we've had to scan them. One of them was what we call a "permanent temporary." There's a permanent device that's outside of the body with a lead going into the body because they need temporary support, and the permanent lead provides greater flexibility. In that 20-patient group, all were safely scanned. Small numbers, but the key message is that with a careful and coordinated program between radiology and cardiology and with careful patient monitoring, we've been able to scan broad groups.

This includes patients with an abandoned lead—that is, a lead that was, at one time, connected to a device and [subsequently] disconnected. The worry there is that the lead could act like an antenna and absorb radiofrequency energy. We carefully monitor the heart rhythm and give the patients the ability to let us know if they're having any symptoms. [As a result,] we've safely scanned roughly a dozen of patients with an abandoned lead.

Dr Watson: Right. You bring up an interesting point with those abandoned leads because we've had some MRI safety issues with patients understanding when they come to us for an MRI and fill out their safety sheet, it used to ask, "Do you have a pacemaker?" A patient who has had their pulse generator explanted can blithely [answer] that they no longer have their pacemaker, without realizing that they still have their implanted leads. We have changed our MRI safety form to say, "Have you ever had a pacemaker?" and set our gain to look for those abandoned leads, which, as you say, can cause a risk in the MRI environment.

Ms. Acker: In 2008, we started to scan MRIs, and over the years, our protocol has changed to incorporate some of these additional findings, like leads that were placed less than 30 days.

Reimbursement Issue: MRI-Conditional vs Non–MRI-Conditional

Dr Kapa: Is there any concern comparing a patient coming with an MRI-conditional device vs a non–MRI-conditional device? Let's say a recently implanted device that is not specifically MRI-conditionally safe?

Dr Friedman: The biggest difference in the clinical realm, I'd say two things—one is for patients who are pacemaker-dependent, it's simpler to scan if their device has been designed specifically to be MRI-safe. There's no question about that.

The real-world challenge we face is that—in the United States, at least—the vast majority of implanted devices are not designed to be MR-conditional. So the vast majority of patients who need a diagnostic MRI scan who have a device have one that isn't built specifically to be MRI scanned. When the battery runs out on their old device and you put in a new one, as Nancy pointed out earlier, it still won't be MR-conditional, most likely, depending on which leads were used. Because we don't change the leads; and if they're not conditional, then the system isn't conditional.

The other issue is financial—and that's undergoing evaluation—but if a patient has a device that's not designed and tested to be MRI-conditional, then the scan may not be reimbursed. That's problematic for patients in whom the diagnosis and treatment is significantly affected by the diagnostic ability of an MRI scan. And that's being actively reviewed by Center for Medicare and Medicaid Services. At Mayo Clinic, we have been scanning these patients without charging when there is a pressing clinical need. That's how we have such a large number, but I think that is something that, on a national level, is undergoing review and needs to be addressed, because the presence of a cardiac implantable electronic device shouldn't be a reason not to have a scan when it can be done safely.

And we're increasingly seeing publications in support of the safety.[1] The key to that safety has been a coordinated effort: that it's not an inadvertent scan of a patient with an implanted device but a preparation ahead of time with specific clinical protocols and monitoring parameters in place so that, for any untoward changes in rhythm, they can react immediately.

Dr Watson: Right. You make a very good point about the reimbursement issue and the fact that we forge ahead and try to do what's right for the patient. One of the sound bites that I've read repeatedly is that patients with cardiac devices statistically will have a 50% to 75% chance of needing an MRI over the lifetime of that device.[2–4] That [indicates] the compelling need to get these types of programs squared away.

Dr Friedman: It's an important issue.

Considering New MRI-Conditional Devices

Dr Kapa: From an implantation perspective, for a patient coming who now needs a device, what are your thoughts as far as always implanting an MRI-conditional device vs using devices that may not be MRI-conditional, since both are on the shelf?

Dr Friedman: That's another excellent question. A number of considerations go into what kind of device you implant. Specific features for individual patients include the type of rate response in pacemakers or defibrillators or the availability of specific antitachycardia pacing. Those devices may not be MR-conditional.

We're also seeing new technologies: subcutaneous defibrillators and leadless pacemakers. Interestingly, of the leadless pacemakers, the one that has been approved is MR-conditional. The one that's undergoing review, likely will be once it's approved. While it's great to have [those] as an option, I would say that many of the devices being implanted today are not MR-conditional. At our institution, more than half are probably not MR-conditional.

Take-Home Message: Scanning Can Be Done

Dr Kapa: Do you have any final comments or anecdotes?

Dr Friedman: I'll wrap up just by saying the following. First, the key take-home messages are: If you have a patient who needs an MRI scan, if they have an implanted device, the scan can be done more that 95% of the time.

The second message is that this requires a coordinated effort between multiple team members in a multidisciplinary team. We need the expertise of cardiac device nurses and MR specialists and their physicist colleagues who know how to set the right imaging parameters, pick the right device, do the right programming, and an entire coordinated multidisciplinary team. We have seen safety [reported by] other centers. My hope is that, increasingly, with those kinds of programs in place and demonstrating safety, a larger number of patients can get the care they need by having these collaborative efforts.

Dr Watson: From the radiology side, compared with if we needed a CT to go to MR, we've had multiple cases where we've been able to make stunning diagnoses [not otherwise possible] for patients, allowing them to receive the appropriate clinical care. Examples, like dural [arteriovenous] AV fistulas, that we found unexpectedly on spine MRIs. Life-changing findings. So, it's been a fabulous program, and we've really enjoyed working with you.

Dr Friedman: It's been a great collaboration.

Ms Dalzell: Yes, we agree.

Dr Kapa: Thanks everyone for these very important insights, and thank you for joining us on on Medscape.