Skip to Main Content

View Public Comments for Magnetic Resonance Imaging (MRI) (CAG-00399R4)

Russo, MD, PhD, FACC, Robert
The Scripps Research Institute

July 4, 2017

Tamara Syrek-Jensen, JD
Director, Division of Items and Devices
Centers for Medicare and Medicaid Services
Office of Clinical Standards and Quality
Coverage and Analysis Group
7500 Security Blvd.
Mail Stop C1-09-06
Baltimore, MD 21244

Re: Formal Request for Reconsideration of the National Coverage Determination (NCD) for Magnetic Resonance Imaging (MRI) to revise the Contraindication for Non-MRIConditional Cardiac Pacemakers in Patients Not Enrolled in a Prospective Clinical Study (Chapter 1, Section 220.2.C.1 in the NCD Manual; other diagnostic tests §1861(s)(3))

Dear Ms. Syrek-Jensen:

On behalf of the undersigned individuals, and professional organizations, we are writing to formally request a revision of the current policy regarding coverage of magnetic resonance imaging (MRI) for Medicare beneficiaries who have a non-MRI-conditional cardiac implantable electrical device (CIED), including permanent pacemakers (PMs) or implantable cardioverter-defibrillator (ICD) (i.e., not approved by the Food and Drug Administration for MRI scanning). As you are aware, it is stated in the current Medicare National Coverage Determinations Manual (NCD), that payment for an MRI examination "will be covered by Medicare when studied in a clinical study under § 1862(a)(1)(E) (consistent with § 1142 of the Act) if the study meets the criteria" included in Decision Memo for Magnetic Resonance Imaging (MRI) (CAG-00399R2)1.

A previous request by one member of our group to the Centers for Medicare & Medicaid Services (CMS) for a revision in the NCD language resulted in Decision Memo CAG- 003999R2, which was published in February 2011. In the Decision Memo it was stated that "CMS believes that the evidence is promising although not yet convincing that MRI will improve patient health outcomes if certain safeguards are in place to ensure that the exposure of the device to an MRI environment adversely affects neither the interpretation of the MRI result nor the proper functioning of the implanted device itself. We believe that specific precautions could maximize benefits of MRI exposure for beneficiaries enrolled in clinical studies designed to assess the utility and safety of MRI exposure." We now feel strongly feel that the weight of the published literature in this field provides "convincing evidence" that, with appropriate precautions, MRI can be performed with minimal risk in Medicare beneficiaries with CIEDs, that the resulting images are interpretable, and that health outcomes for Medicare beneficiaries are improved. This letter outlines our rationale.

A. MagnaSafe Registry

A multicenter study with the goal of determining the frequency of cardiac device-related clinical events and device setting changes among patients with a non-MRI-conditional device who underwent nonthoracic MRI at a field strength of 1.5 tesla began enrollment in 2009 (The MagnaSafe Registry2). The MagnaSafe Registry experimental protocol was written after consultation with personnel at the Center for Devices and Radiological Health of the US Food and Drug Administration (FDA), and an investigational device exemption (IDE) was obtained for the purpose of data collection and adverse event reporting. The MagnaSafe Registry continued after publication of Decision Memo CAG-003999R2. It was subsequently noted in Decision Memo CAG-00399R3 (July 2011), that in the opinion of the representatives of the CMS Coverage and Analysis Group, that they "believed that the MagnaSafe registry appears to meet the CED (coverage with evidence development) requirement."3

The results of the MagnaSafe Registry were published in February 2017.4 In this study, a total of 1000 pacemaker cases and 500 ICD cases with non-MRI-conditional systems were enrolled in centers in the United States. The MRI examinations were clinically indicated in the opinion of the patient's ordering physician, and MRI scans of the chest (thoracic MRI including cardiac imaging and thoracic spine imaging) were excluded from study entry. Cardiac devices were interrogated before, and after MRI with the use of a standardized protocol, and the devices were appropriately reprogrammed before the MRI examination. The primary end points of the study were death, generator or lead failure, induced arrhythmia, loss of capture for pacing-dependent patients or electrical reset during the examination. The secondary end points were pre-determined changes in device settings.

In this study, no deaths, lead failures, losses of capture in pacing-dependent patients, or ventricular arrhythmias occurred during MRI. One ICD generator could not be interrogated after MRI and required immediate replacement; the device had not been appropriately programmed per protocol before the MRI. However, a similar event may have occurred with an MRI-conditional ICD, had such a device not been appropriately programmed prior to the MRI examination. Six cases of self-terminating atrial fibrillation or flutter and six cases of partial electrical reset were observed. Changes in lead impedance, pacing threshold, battery voltage, and P-wave and R-wave amplitude exceeded prespecified levels in a small number of cases. Repeat MRI examinations were not associated with an increase in adverse events. This study demonstrated that patients with a non-MRI-conditional pacemaker or ICD could undergo clinically indicated nonthoracic MRI at 1.5 tesla, without risk to the patient or device, when the patient was appropriately screened, and the device was reprogrammed in accordance with the prespecified protocol.

B. Additional Literature

Prior to the publication of the MagnaSafe Registry results, Nazarian, et al5, published "A Prospective Evaluation of a Protocol for Magnetic Resonance Imaging of Patients With Implanted Cardiac Devices." The purpose of this study was to define the safety of a protocol for MRI at a field strength of 1.5 T in patients with an implanted cardiac device. In this study, patients with either a non-MRI-conditional pacemaker or defibrillator underwent a total of 555 MRI scans; 18% of these scans included the heart or thoracic spine.

The results of this study included 3 cases in which the device reverted to a transient back-up programming mode without long-term effects, and ventricular sensing and atrial and ventricular lead impedances were reduced by a small amount immediately after MRI. At long-term follow-up, decreased ventricular sensing, decreased ventricular lead impedance, increased ventricular capture threshold, and decreased battery voltage were noted. However, the observed changes did not require device reprogramming or replacement.

In addition to the 2,055 cases enrolled in the two studies noted above, an additional 1,888 cases were reported in several smaller studies examining the risk associated with MRI in patients with a non-MRI-conditional implanted device4-24 (Table 1). In these studies, the authors reported varying effects on cardiac device settings. Overall, MR scanning was performed safely; electrical resets were rarely seen and were successfully reprogrammed after the procedure. Pacing thresholds were noted to increase and decrease, but rarely required a change in programming.

It was also stated in Decision Memo CAG-003999R2 that "CMS believes that the evidence is promising although not yet convincing that MRI will improve patient health outcomes if certain safeguards are in place to ensure that the exposure of the device to an MRI environment" does not adversely affect "the interpretation of the MRI result..." Recently, Mukai et al25 presented "Does the presence of an implanted cardiac device adversely affect the image quality of clinically indicated magnetic resonance imaging at 1.5t?" The authors reported the imaging results of 1000 consecutive clinically indicated MRI examinations at 1.5 tesla performed in 569 patients with an implanted pacemaker of defibrillator. Device-related imaging artifacts were reported in 1.0% of non-cardiac scans, but none of the 976 non-cardiac MRI studies "contained an artifact that adversely affected image quality with the requirement for an alternate imaging modality."

Lastly, Strom et al42 described a case series of 189 MRI examinations performed in 123 patients. In this series 98.4% of scans were deemed to be interpretable. Using a pre-specified adjudication system for determining the clinical utility of MRI, nearly 80% of MRI examinations that met the requirement of an interpretable scan, also led to a change in treatment or diagnosis or guided a subsequent procedure.

C. Professional Society Guidelines

In May 2017 Indik, et al26, published the "2017 Heart Rhythm Society (HRS) expert consensus statement on magnetic resonance imaging and radiation exposure in patients with cardiovascular implantable electronic devices," which is intended to help health care providers involved in the care of adult (and pediatric) patients with cardiac implantable electronic devices who are to undergo MRI. This document addresses the recommended procedures for MRI in patients with MRI-conditional and non-MRI-conditional pacemakers and implanted defibrillators. Regarding the management of patients with an non-MRI-conditional device undergoing an MRI examination, it is stated that: "It is reasonable for patients with an non-MRI-conditional system to undergo MR imaging if there are no fractured, epicardial, or abandoned leads; the MRI is the best test for the condition; and there is an institutional protocol and a designated responsible MR physician and CIED physician."

It should be noted that the HRS consensus statement26 was developed in collaboration with and endorsed by the American College of Cardiology (ACC), American College of Radiology (ACR; endorsement pending), American Heart Association (AHA), American Society for Radiation Oncology (ASTRO), Asia Pacific Heart Rhythm Society (APHRS), European Heart Rhythm Association (EHRA), Japanese Heart Rhythm Society (JHRS), Pediatric and Congenital Electrophysiology Society (PACES), Brazilian Society of Cardiac Arrhythmias (SOBRAC), and Latin American Society of Cardiac Stimulation and Electrophysiology (SOLAECE) and in collaboration with the Council of Affiliated Regional Radiation Oncology Societies (CARROS).

D. Role of MRI in Clinical Medicine

Over the past decade, magnetic resonance imaging has become the imaging modality of choice for the evaluation of many, if not most diseases of the brain, spinal cord, and musculoskeletal system. Approximately 2-3 million people in the United States (many of them Medicare beneficiaries) have a non-MRI-conditional cardiac pacemaker or implantable cardioverter-defibrillator (ICD).5 It is predicted that at least half of these patients will have a clinical indication for MRI during their lifetime after device implantation.27 Improved access to MRI for Medicare beneficiaries with non-MRI-conditional pacemakers and ICDs will improve health outcomes. For example, MRI has been proven to be superior to computed tomography (CT) for the evaluation of

  • Acute ischemic stroke28
  • Detection of multiple sclerosis lesions.29
  • Acute intracerebral hemorrhage30
  • Detection of dysplastic hepatic nodules and early hepatocellular carcinoma31,32
  • Whole-body imaging in patients with metastatic breast cancer33

In addition, Appropriateness Criteria from the American College of Radiology34 rates MRI higher than CT for clinical decision-making in patients for:

  • Breast Cancer Screening in high-risk women with a BRCA gene mutation and their untested first- degree relatives, women with a history of chest irradiation between the ages of 10-30, and women with 20% or greater lifetime risk of breast cancer
  • Abdominal imaging with a liver lesion for initial characterization of an indeterminate, >1 cm lesion on initial imaging with ultrasound, and a normal liver
  • Brain imaging in the evaluation of Alzheimer's disease, frontotemporal dementia, dementia with Lewy bodies, vascular dementia, Creutzfeld-Jakob or other prion mediated dementia, normal pressure hydrocephalus, neurodegeneration with brain iron accumulation, and Parkinsonian syndrome
  • Brain imaging for patients with acute or chronic headache
  • Brain imaging for patients with single or multiple focal neurologic deficits, of subacute onset, with progressive or fluctuating symptoms
  • Brain imaging for patients with a suspected, or previously treated primary or metastatic brain malignancy
  • Brain imaging for patients with seizures and epilepsy for the purpose of surgical planning, and for the evaluation of new onset seizures with or without head trauma.
  • Brain imaging for patients with acute or subacute ataxia without head trauma
  • Imaging of the orbits for patients with sudden non-traumatic onset of painless or painful visual loss.
  • Imaging of the lumbar spine with acute, subacute, or chronic low back pain or radiculopathy

Therefore, the use of CT rather than MRI for patients with a non-MRI-conditional pacemaker or ICD may lead to an incorrect diagnosis, and possibly inappropriate, or incomplete therapy in many disease states.

In 2011, the FDA approved the first MRI-conditional pacemaker generator-and-lead for marketing in the United States (Revo MRI SureScan pacing system, Medtronic, Inc.).35,36 Since that time, at least 13 generator-and-lead systems from four manufacturers have received FDA approval as an MRI-conditional system. Although it has been suggested that previously implanted non-MRI-conditional generators and leads may be removed and then replaced to allow for MRI, the potential risks from such procedures are much greater than those associated with MRI with an implanted non-MRI-conditional device. The rate of major complications among patients undergoing generator replacement with or without the placement of an additional transvenous lead was 4 to 15% in a prospective registry.37 In addition, single-center and multicenter studies have shown a rate of major complications (including death and tamponade) associated with elective laser-assisted lead extraction that is in the range of 0.4 to 2%.38-41 These results strongly suggests that device removal and replacement are unlikely to be safer than proceeding with scanning for patients with a non-MRI-conditional pacemaker or an ICD who require an MRI examination.

E. Requested Coverage Modifications

We feel the results of the MagnaSafe registry as well as the many studies that have examined the risk of MRI in patients with a non-MRI-conditional device, and have been published since the last request to change the NCD, provide sufficient evidence to demonstrate that MRI at a field strength of 1.5 tesla can be performed with minimal risk for patients who have a non-MRI-conditional pacemaker or ICD, when the patients are properly monitored and the device is appropriately reprogrammed before the examination following a protocol designed to minimize adverse events to patients and their devices.

Thus, we request that the current Medicare National Coverage Determination language be modified to allow coverage for Medicare beneficiaries with non-MRI-conditional pacemakers or ICDs who undergo MRI at a field strength of 1.5 tesla without the requirement for entry into a research study, when there is a strong clinical indication for the MRI examination, with no acceptable alternative imaging modality, and the patient is appropriately monitored. We further request that the language in section 220.2, section C.1; other diagnostic tests §1861(s)(3) in the National Coverage Determination be revised to remove the requirement for enrollment "in clinical studies designed to assess the utility and safety of MRI exposure" and to remove the requirements for coverage with evidence development. We suggest that the language be revised to read as follows:

"An MRI procedure for patients (Medicare beneficiaries) with a non-MRI-conditional pacemaker or ICD will be covered if the examination is performed in accordance with the 2017 Heart Rhythm Society expert consensus statement on magnetic resonance imaging and radiation exposure in patients with cardiovascular implantable electronic devices (Indik, et al26)." The recommendations in that document include the following*:

  1. It is reasonable for patients with a non-MRI-conditional CIED system to undergo MR imaging if there are no fractured, epicardial, or abandoned leads; MRI is the best test for the condition; and there is an institutional protocol and a designated responsible MR physician and CIED physician.
  2. It is reasonable to perform an MR scan immediately after implantation of a lead or generator of a non-MRI-conditional CIED system if clinically warranted.
  3. For patients with a non-MRI-conditional CIED, it is reasonable to perform repeat MRI when required, without restriction regarding the minimum interval between imaging studies or the maximum number of studies performed.
  4. It is recommended for the patient with a non-MRI-conditional CIED that device evaluation be performed immediately pre- and post-MRI with documentation of pacing threshold(s), P- and R-wave amplitude, and lead impedance using a standardized protocol.
  5. A defibrillator/monitor (with external pacing function) and a manufacturer-specific device programming system should be immediately available in the holding area adjacent to the MR scanner room while an Non-MRI-conditional CIED is reprogrammed or imaging.
  6. It is recommended that continuous MR conditional ECG and pulse oximetry monitoring be used while a Non-MRI-conditional CIED is reprogrammed for imaging.
  7. It is recommended that personnel with the skill to perform advanced cardiac life support, including expertise in the performance of CPR, arrhythmia recognition, defibrillation, and transcutaneous pacing, accompany the patient with an Non-MRI-conditional CIED for the duration of time the patient's device is reprogrammed, until assessed and declared stable to return to unmonitored status.
  8. For patients with a Non-MRI-conditional CIED who are pacing-dependent (PM or ICD), it is recommended that: a) Personnel with the skill to program the CIED be in attendance during MR scanning. b) A physician with the ability to establish temporary transvenous pacing be immediately available on the premises of the imaging facility. c) A physician with the ability to direct CIED programming be immediately available on the premises of the imaging facility.
  9. For patients with a non-MRI-conditional CIED who are not pacing-dependent, it is recommended that: a) Personnel with the skill to program the CIED be available on the premises of the imaging facility. b) A physician with the ability to direct CIED programming be available on the premises of the imaging facility.
  10. It is recommended that for the patient with a non-MRI-conditional CIED who is pacing-dependent to program their device to an asynchronous pacing mode with deactivation of advanced or adaptive features during the MRI examination, and the pacing rate should be selected to avoid competitive pacing.
  11. All tachyarrhythmia detections for patients with an ICD should be disabled prior to MRI.
  12. The MR-responsible physician who is accountable for overseeing the safety of the MRI environment, including the administration of any medication and/or contrast agents (if applicable), should be made aware of the presence of a patient with an Non-MRI-conditional CIED.
  13. It is recommended that ECG and pulse oximetry monitoring be continued until baseline or until other clinically appropriate CIED settings are restored for patients with a Non-MRI-conditional CIED.
  14. All resuscitative efforts and emergency treatments that involve the use of a defibrillator/monitor, device programming system, or any other MRI-unsafe equipment should be performed after moving the patient outside of Zone 4.
  15. For a patient with a Non-MRI-conditional CIED who is not pacing-dependent, it is reasonable to program their device to either a non-pacing mode (OVO/ODO) or to an inhibited mode (DDI/VVI), with deactivation of advanced or adaptive features during the MRI examination.
  16. It is reasonable to program patients with a Non-MRI-conditional CRT device who are not pacing-dependent to an asynchronous pacing mode (VOO/DOO) with deactivation of advanced or adaptive features during the MRI examination, and with a pacing rate that avoids competitive pacing.
  17. For patients with a Non-MRI-conditional CIED, it is reasonable to schedule a complete follow-up CIED evaluation within 1 week for a pacing lead threshold increase ≥ 1.0 V, P-wave or R-wave amplitude decrease ≥ 50%, pacing lead impedance change ≥ 50 Ω, and high-voltage (shock) lead impedance change .5 Ω, and then as clinically indicated.

(*The recommendations #1-17 pertaining to "Recommendations for the Decision to Perform an MRI on Patients with an MR Nonconditional CIED", are provided in the section above without alteration from Indik, et al.26)

Thank you for your consideration of our request. If you have any questions or require additional information, please feel free to contact either Dr. Russo by phone at 858-886-7595, or by email at; and/or Dr. Kramer by email at


Robert J. Russo, MD, PhD
The Scripps Research Institute

Daniel B. Kramer, MD, MPH
Beth Israel Deaconess Medical Center
Harvard Medical School, Boston

Cardiovascular Diseases

Jeffrey L. Anderson, MD
Intermountain Medical Center Heart Institute

Robert W.W. Biederman, M.D.
Allegheny General Hospital, Pittsburgh

Noel G. Boyle, MD, PhD
University of California, Los Angeles

Paul Friedman, MD
Mayo Clinic

J. Rod Gimbel, MD, FACC, FHRS
Clinical Cardiac Electrophysiologist

Ulrika Birgersdotter-Green, MD
Unversity of California, San Diego

Henry Halperin, MD, MA
Johns Hopkins University School of Medicine

Julia H. Indik, MD PhD
University of Arizona

Aaron S Kesselheim, MD, JD, MPH
Brigham and Women's Hospital
Harvard Medical School, Boston

Rachel Lampert, MD
Yale University School of Medicine

Harold Litt, MD
University of Pennsylvania

Edward T. Martin, M.D.
Oklahoma Heart Institute, Tulsa

John Mandrola, MD
Baptist Health, Louisville, KY.

Saman Nazarian, MD, PhD
The University of Pennsylvania Perelman School of Medicine

Matthew Reynolds, MD, MSc
Lahey Hospital and Medical Center
Baim Institute for Clinical Research


Emanuel Kanal, MD
University of Pittsburgh Medical Center

Pamela K. Woodward, MD
Washington University School of Medicine

Endorsement requested but not yet received:

The American College of Cardiology
Representative; To be determined

The American Heart Association
Representative; To be determined

The Heart Rhythm Society
Representative; To be determined

The American College of Radiology
Representative; To be determined


2. Russo RJ. Determining the risks of clinically indicated nonthoracic magnetic resonance imaging at 1.5 T for patients with pacemakers and implantable cardioverter-defibrillators: rationale and design of the MagnaSafe Registry. Am Heart J 2013; 165:266-72.
4. Russo RJ, Costa HS, Silva PD, et al. Assessing the risks associated with MRI in patients with a pacemaker or defibrillator. N Engl J Med 2017;376:755-764.
5. Nazarian S, Hansford R, Roguin A, et al. A prospective evaluation of a protocol for magnetic resonance imaging of patients with implanted cardiac devices. Ann Intern Med 2011;155:415-424.
6. Martin ET, Coman JA, Shellock FG, et al. Magnetic resonance imaging and cardiac pacemaker safety at 1.5-Tesla. J Am Coll Cardiol 2004;43(7):1315-1324.
7. Sommer T, Naehle CP, Yang A, et al. Strategy for safe performance of extrathoracic magnetic resonance imaging at 1.5 tesla in the presence of cardiac pacemakers in non-pacemaker-dependent patients: a prospective study with 115 examinations. Circulation 2006;114(12):1285-1292.
8. Nazarian S, Roguin A, Zviman MM, et al. Clinical utility and safety of a protocol for noncardiac and cardiac magnetic resonance imaging of patients with permanent pacemakers and implantable-cardioverter defibrilltors at 1.5 tesla. Circulation 2006;114(12):1277-1284.
9. Pulver AF, Puchalski MD, Bradley DJ, et al. Safety and imaging quality of MRI in pediatric and adult congenital heart disease patients with pacemakers. Pacing Clin Electrophysiol 2009;32(4):450-456.
10. Burke PT, Ghanbari H, Alexander PB, et al. A protocol for patients with cardiovascular implantable devices undergoing magnetic resonance imaging (MRI): should defibrillation threshold testing be performed post-(MRI). J Interv Card Electrophysiol 2010;28(1):59-66.
11. Buendia F, Sanchez-Gomez JM, Sancho-Tello MJ, et al. Nuclear magnetic resonance imaging in patients with cardiac pacing devices. Rev Esp Cardiol 2010; 63(6):735-739.
12. Cohen JD, Costa HS, Russo RJ. Determining the risks of magnetic resonance imaging at 1.5 tesla for patients with pacemakers and implantable cardioverter defibrillators. Am J Cardiol 2012;110(11):1631-1636.
13. Strach K, Naehle CP, Muhlsteffen A, et al. Low-field magnetic resonance imaging: increased safety for pacemaker patients? Europace 2010;12(7):952-960.
14. Muehling OM, Wakili R, Greif M, et al. Immediate and 12 months follow up of function and lead integrity after cranial MRI in 356 patients with conventional cardiac pacemakers. J Cardiovasc Magn Reson 2014;16:39.
15. Junttila MJ, Fishman JE, Lopera GA, et al. Safety of serial MRI in patients with implantable cardioverter defibrillators. Heart 2011;97(22):1852-1856.
16. Boilson BA, Wokhlu A, Acker NG, et al. Safety of magnetic resonance imaging in patients with permanent pacemakers: a collaborative clinical approach. J Interv Card Electrophysiol 2012;33(1):59-67.
17. Del Ojo JL, Moya F, Villalba J, et al. Is magnetic resonance imaging safe in cardiac pacemaker recipients? Pacing Clin Electrophysiol 2005; 28(4):274-278.
18. Gimbel JR, Bailey SM, Tchou PJ, et al. Strategies for the safe magnetic resonance imaging of pacemaker-dependent patients. Pacing Clin Electrophysiol 2005;28(10):1041-1046.
19. Mollerus M, Albin G, Lipinski M, et al. Cardiac biomarkers in patients with permanent pacemakers and implantable cardioverter-defibrillators undergoing an MRI scan. Pacing Clin Electrophysiol 2008;31(10):1241-1245.
20. Mollerus M, Albin G, Lipinski M, et al. Magnetic resonance imaging of pacemakers and implantable cardioverter-defibrillators without specific absorption rate restrictions. Europace 2010;12(7):947-951.
21. Friedman HL, Acker N, Dalzell C, et al. Magnetic resonance imaging in patients with recently implanted pacemakers. Pacing Clin Electrophysiol 2013;36(9):1090-1095.
22. Higgins JV, Sheldon SH, Watson RE, et al. "Power-on resets" in cardiac implantable electronic devices during magnetic resonance imaging. Heart Rhythm 2015;12(3):540-544.
23. Naehle CP, Strach K, Thomas D, et al. Magnetic resonance imaging at 1.5-T in patients with implantable cardioverter-defibrillators. J Am Coll Cardiol 2009;54(6):549-555.
24. Higgins JV, Gard JJ, Sheldon SH, et al. Safety and outcomes of magnetic resonance imaging in patients with abandoned pacemaker and defibrillator leads. Pacing Clin Electrophysiol 2014;37(10):1284-1290.
25. Mukai K, Costa H, and Russo, RJ. Does the presence of an implanted cardiac device adversely affect the image quality of clinically indicated magnetic resonance imaging at 1.5t? J Cardio Mag Res 2015;17(Suppl 1):241
26. 2017 HRS expert consensus statement on magnetic resonance imaging and radiation exposure in patients with cardiovascular implantable electronic devices. Washington, DC: Heart Rhythm Society, May 11, 2017 (
27. Kalin R, Stanton MS. Current clinical issues for MRI scanning of pacemaker and defibrillator patients. Pacing Clin Electrophysiol 2005;28:326-328
28. Chalela JA, Kidwell CS, Nentwich LM, et al. Magnetic resonance imaging and computed tomography in emergency assessment of patients with suspected acute stroke: a prospective comparison. The Lancet. 2007;369(9558):293-298.
29. Ge Y. Multiple sclerosis: the tole of MR imaging. AJNR American Journal of Neuroradiology. 2006;27(6):1165-1176.
30. Kidwell CS, Chalela JA, Saver JL, et al. Comparison of MRI and CT for detection of acute intracerebral hemorrhage. JAMA. 2004;292(15):1823-1830.
31. Kim BR, Lee JM, Lee DH, et al. Diagnostic Performance of Gadoxetic Acid-enhanced Liver MR Imaging versus Multidetector CT in the Detection of Dysplastic Nodules and Early Hepatocellular Carcinoma. Radiology. 2017 Jun 13:162080. [Epub ahead of print]
32. Park MJ, Hong N, Han K, et al. Use of Imaging to Predict Complete Response of Colorectal Liver Metastases after Chemotherapy: MR Imaging versus CT Imaging. Radiology. 2017 Mar 22:161619.
33. Kosmin M, Makris A, Joshi PV, et al. The addition of whole-body magnetic resonance imaging to body computerised tomography alters treatment decisions in patients with metastatic breast cancer. Eur J Cancer. 2017 May;77:109-116.
34. American College of Radiology. Appropriateness criteria. 2017 (https://acsearch.acr .org/list).
35. Wilkoff BL, Bello D, Taborsky M, et al. Magnetic resonance imaging in patients with a pacemaker system designed for the magnetic resonance environment. Heart Rhythm 2011; 8: 65-73.
36. Gimbel JR, Bello D, Schmitt M, et al. Randomized trial of pacemaker and lead system for safe scanning at 1.5 Tesla. Heart Rhythm 2013; 10: 685-91.
37. Poole JE, Gleva MJ, Mela T, et al. Complication rates associated with pacemaker or implantable cardioverter-defibrillator generator replacements and upgrade procedures: results from the REPLACE registry. Circulation 2010;122:1553-1561
38. Hauser RG, Katsiyiannis WT, Gornick CC, Almquist AK, Kallinen LM. Deaths and cardiovascular injuries due to device-assisted implantable cardioverter-defibrillator and pacemaker lead extraction. Europace 2010;12:395-401.
39. Jones SO IV, Eckart RE, Albert CM, Epstein LM. Large, single-center, single-operator experience with transvenous lead extraction: outcomes and changing indications. Heart Rhythm 2008;5:520-525.
40. Wazni O, Epstein LM, Carrillo RG, et al. Lead extraction in the contemporary setting: the LExICon study: an observational retrospective study of consecutive laser lead extractions. J Am Coll Cardiol 2010;55:579-586.
41. Wilkoff BL, Byrd CL, Love CJ, et al. Pacemaker lead extraction with the laser sheath: results of the Pacing Lead Extraction with the Excimer Sheath (PLEXES) trial. J Am Coll Cardiol 1999;33:1671-1676.
42. Strom JB, Whelan JB, Shen C, et al. Safety and utility of magnetic resonance imaging in patients with cardiac implantable electronic devices. Heart Rhythm. 2017, S1547-5271(17)30416-2. [Epub ahead of print]

Table 1: Literature regarding the management of patients with an non-MRI-conditional device who are undergoing MRI
(Adapted with permission from Indik, JH, et al26)
Study author and Year Study type and size Inclusion criteria Endpoints Findings Outcomes* Statistics† Limitations
Martin et al6

Single center, prospective cohort

N = 54 patients, 62 MR scans

Included cardiac, vascular and general MR studies, no restrictions on PM type but PM-dependent excluded

Pacing threshold post-MRI evaluated for "any change" or "any significant change"

Any significant change defined as change >1 voltage or pulse width increment/decrement

A total of 9.4% of leads had significant changes, with 1.9% requiring change in programmed output, but unrelated to cardiac chamber, anatomic location, peak SAR, or time from implant to MR scan No adverse outcomes, patient symptoms and ECG changes minor and did not require cessation of MRI Logistic regression for peak SAR, chi-squared or Fisher exact for 2¡Ñ2 contingency testing Single center, only immediate post-MRI PM evaluation performed
Sommer et al7

Single center, prospective cohort

N = 82 patients,

115 MR scans

PM patients who were not dependent, MR scans not of thoracic region, urgent need for MR scan,

Medtronic PMs manufactured 1993-2004, with stable device parameters

Change in pacing threshold clinically significant if 3 1V

Pacing threshold increased pre- to post-MRI (P = .017), and clinically significant in 3.1% of leads (95% CI 1.1-6.6%), and in 2 leads increase in threshold detected at follow-up at 3 months

Electrical reset occurred after 7 scans

Troponin increased in 4 of 114 scans, and in one case rise of troponin associated with significant change in threshold, but overall no significant increase in troponin (P = .0693)

No inhibition of pacing or arrhythmia observed and scans performed safely

PMs with electrical reset all programmed back to pre-scan parameters

No leads required change in output to maintain function

Mixed repeated-measures regression analysis of threshold and impedance data, with covariates for cardiac chamber, timing of evaluation (pre, post, 3-month follow-up)

Single center

Medtronic only

Nazarian et al8

Single center, prospective cohort

N = 55 patients,

31 PM

24 ICD,

68 MR scans

Patients included if no imaging alternative and could be pacing-dependent

Excluded if < 6 weeks from implant, nontransvenous leads, abandoned leads

Change in PM parameters from pre- to immediate post- and long-term follow-up

N = 12 pacing-dependent

No inappropriate inhibition or pacing

No significant differences in amplitude, impedance, threshold from pre-scan to immediate post-scan or to long term f/u (median 99 days) Paired Student t test to compare immediate and long term parameters Single center
Pulver et al9

Single center, prospective case series of adult and pediatric patients with congenital heart disease

N = 8 patients, with N = 11 MR scans

Could have epicardial leads

Not pacing-dependent and no abandoned leads


Lead parameters

Average age 16.5 ± 9.2 years, and 5 under age 16

No inappropriate pacing or significant change in parameters noted pre- to post-MR scan

9 epicardial leads included

Exams performed safely

Long-term follow-up data available on 6 patients with no clinically important changes seen

Paired t tests to compare pacing parameters pre- and post-MR Small case series
Burke et al10

Single center, prospective cohort

N = 38 patients, 92 MR scans

Indication for MR would result in significant clinical impact Device parameters including DFTs immediate post MR and at 3-month follow up N=13 PM-dependent, N = 11 not PM-dependent, N = 10 ICD patients, N = 4 CRT patients

No device circuitry damage, programming alterations, no electrical resets, inappropriate shocks, failure to pace or changes in sensing, pacing, or defibrillation threshold, including patients with multiple MR scans

No change in device parameters at 3-month follow-up

Paired t test and Wilcoxon rank sum test Single center
Buendia et al11

Single center, prospective cohort

N = 33 patients

PPM 28


MR clinically essential


Lead parameters

N = 28 with PMs, N = 5 with ICDT

Noted: temporary communication failure in two patients; Sensing errors during imaging in two patients

Safety signal generated in one PM at the maximum magnetic resonance frequency and output level

No technical restrictions on imaging or any permanent change in CIED performance, no clinical complications   Small case series
Cohen et al12

Single center, retrospective cohort that underwent MR and prospective (control) cohort that did not undergo MR

Retrospective cohort: N = 109 patients, with N = 125 clinically indicated MR scans

Prospective cohort: N = 50 patients with CIED

All patients with permanent CIEDs who underwent clinically necessary MR scans from 2006-2009

Control group recruited from 2008-2009

Underwent two interrogations one hour apart

Primary endpoints: death during MR, device or lead failure requiring immediate replacement, induced atrial or ventricular arrhythmias during MR, loss of PM capture, electrical reset

Secondary endpoints: battery voltage decrease of 3 0.4 V, pacing lead threshold increase of 3 0.5 V at 0.4 ms pulse width, P-wave amplitude decrease 3 50%, R-wave amplitude decrease 3 25%, lead impedance change 3 50 W, high voltage lead impedance change 3 3W

Pacer dependence: 27% in MR group, 16% in control group

No significant change between MR and control groups for battery voltage, P-wave amplitude, R-wave amplitude, or high voltage impedance

Small mean decrease in LV threshold in MR group and small mean increase in control group noted

Significant difference seen in MR group vs. control for lead impedance (P = .01), but not clinically important

No deaths, device failures, generator/lead replacements, loss of capture, or electrical reset Linear mixed model analyses to compare MR and control groups for CIED parameters, adjusting for type of device and PM dependence Retrospective MR cohort, single center
Strach et al13

Single center, prospective cohort

N = 114 patients with scans performed at 0.2 Tesla, including PM-dependent and abandoned leads

Urgent clinical need for MR scan

Implants at least 3 months prior to scan with stable pacing parameters

Excluded ICD

Evaluation pre- and post-MR No induction of arrhythmias or inhibition of pacing, and no statistically significant changes in lead impedance, pacing threshold, or battery voltage. In no patient was a pacing threshold over 0.5 V observed No adverse effects; MR at low field strength appeared to be safe and feasible Wilcoxon signed rank test to compare pre- and post-MR parameters Number of patients with abandoned leads or details not provided
Nazarian et al5

Single center, prospective cohort

N = 438 patients, with N = 555 MR scans

Consecutively enrolled from 2003-2010

Included PM-dependent patients implanted >6 weeks prior to MR scan


Excluded abandoned or epicardial leads

Excluded ICD patients who were pacing-dependent

Device function at immediate and long-term follow up, safety

Power-on reset occurred in 0.7% thoracic imaging, associated with decreased (compared to nonthoracic) acute RV (P = .005) and long-term RV R-wave amplitude (P = .009)

Small decreases in device parameters seen but not clinically important immediate post-MR: RV amplitude (P <.001), atrial impedance (P <.001), RV impedance (P <.001), LV impedance (P = .002), battery voltage (P <.001)

Small decreases in device parameters but not clinically important in long-term follow-up: RV amplitude (P = .004), RV impedance (P = .044), RV threshold (P = .12), battery voltage (P <.001)

MR performed safely

Changes in device variables did not require device revision or reprogramming

Wilcoxon signed rank test Single center
Muehling et al14

Single center prospective

N = 356 patients, cranial MRI

PM patients needing urgent cranial MRI, included pacing-dependent patients, PMs implanted at least 2 months prior to scan; excluded epicardial or fractured lead; enrolled from 2004-2012

Evaluation of pacing parameters pre-, immediate, post-MR scan and follow-up at 2 weeks, and 2,6, and 12 months after scan

Measurement of troponin 12 hours post-scan

No immediate or late PM dysfunction, no increase in troponin within 12 hours

Programmed parameters unchanged, data for threshold, sensing, impedance did not change significantly, with 19 patients having a maximum increase of 0.4 V in threshold seen

No significant changes in device parameters (sensing, impedance or pacing capture threshold) up to 12 months

Paired Wilcoxon rank sum test for continuous variables, Kruskal-Wallis for categorical variables

Pre- and post-scans compared by ANOVA

Single center,

Cranial MRI only

PM patients only

Russo et al4

Multicenter prospective registry

N = 1000 PM cases (848 patients), and N = 500 (428 patients) ICD cases

Nonthoracic MR scans at 1.5 T

Excluded patients with CIEDs implanted before 2002

Excluded ICD patients that were pacing-dependent

Primary outcomes: death, generator or lead failure that required immediate replacement, loss of capture, new onset arrhythmia during scan, partial or full electrical reset

Secondary outcomes: decrease in battery voltage 3 0.4 V, increase in pacing threshold 3 0.5 V at 0.4 ms, decrease in P-wave 3 50%, decrease in R-wave 3 25%, increase/decrease in lead impedance 350 Ω, increase/decrease in shock impedance3 3 Ω

P wave: ≥50% decrease in .9% of PMs, 0.3% of ICDs

R wave: ≥50% decrease in no PMs and 0.2% of ICDs

Pacing threshold: ≥0.5 V in 0.7% of PMs, 0.8% of ICDs

Lead impedance: ≥50 Ω in 3% of PMs, 4% of ICDs

Repeat scanning performed in 22.6% of PMs and 18% of ICDs, with median interval between scans of 153 days for PM patients, 91 days for ICD patients

No deaths, lead failures, losses of capture or ventricular arrhythmias during MRI

5 patients had atrial fibrillation and one atrial flutter during MRI

One ICD generator required replacement because it had not been programmed appropriately for scanning

6 partial electrical resets

95% CIs calculated for observed proportions of binary outcomes Thoracic MRI excluded, also only small number of CRT devices
Junttila et al15

Single center, prospective case series

N = 10 ICD patients who underwent 3 serial cardiac MR scans

Excluded pacing-dependent patients Evaluation of device parameters pre- and post-MR and at follow-up and 3, 6, and 12 months Median follow-up 370 days

No adverse effects with serial MR scans

No differences in pacing capture threshold, lead or high voltage lead impedance, or battery voltage, and no ICD dysfunction

Student t test and Mann-Whitney test Small series, single center; troponin/cardiac biomarkers not measured
Boilson et al16

Single center, prospective cohort

N = 32 patients with 46 MR scans

Not pacing-dependent, with PM (excluded ICD), implanted at least 90 days prior to scan

Safety, lead parameters,

cardiac enzymes

No significant change in battery voltage, sensed P/R waves, pacing thresholds, impedance immediately after MR or at 1 month follow-up

No increase in cardiac enzymes

PVCs noted in one patient

Power-on reset occurred in 5 scans (5 patients), more frequent with Medtronic Kappa

No adverse clinical events

Fisher exact test, Pearson chi-squared tests for categorical values

ANOVA for continuous variables

MR scan of head (N = 35) and spine (12 cervical, 7 thoracic, 5 lumbar)
Del Ojo17 et al

Prospective, single center, case series

N = 13 patients, undergoing MR scan at 2 Tesla


Not pacing-dependent


Lead parameters

No significant differences in sensing, stimulation, threshold, or impedance pre- and post-MR scan No PM inhibition, asynchronous pacing, or inappropriate rapid pacing occurred Student t test Small case series, St. Jude PM only
Gimbel et al18

Prospective cohort with substudy of PM-dependent patients

N = 10 patients with 11 MR scans from 1994-2004


No chest or abdominal MR scans


Lead Parameters pre-, post-MR scan, and at 3 months

No PM malfunction, pauses, or rapid pacing

No power-on resets

No clinically important change in pacing parameters

One patient with a Y adaptor in system Not provided Small series
Mollerus et al19

Single center, prospective cohort

N = 37 patients with 40 MR scans

Not pacing-dependent


Any anatomic body region and no peak specific absorption rate (SAR) limit

Device evaluation pre- and post-MR scan; troponin and myoglobin levels pre- and 6-12 hours post-MR scan

Troponin levels unchanged post-MR scan

No significant change in atrial or ventricular pacing thresholds noted

Median SAR 2.4 W/kg

MR scan performed safely and no change in cardiac biomarkers Wilcoxon rank sum test

Single center, small cohort

Excluded pacing-dependent patients

No long-term follow-up

Mollerus et al20

Single center, prospective cohort

N = 103 patients, with 127 MR scans

Not pacing-dependent

PM, ICD, or CRT, implanted at least 6 weeks prior to scan

No restriction on SAR

Device evaluation pre- and post-MR scan and followed for at least 3 months

Median peak SAR measurements of 2.5 W/kg

Pre- and post-scan pacing thresholds unchanged

Sensed RV amplitudes (P <.00001) and lead impedances (RA, RV) (P <.0001) decreased

One patient with device reset

One ICD had arrhythmia log erased during scan

No significant study-related events seen at 3-month follow-up

Paired Wilcoxon rank sum test for continuous variables, Kruskal-Wallis test for categorical values Single center; excluded pacing-dependent
Friedman et al21

Prospectively collected single-center cohort with retrospective analysis of patients with or without recently implanted leads

N = 171 patients with 219 scans, of which 8 had recently implanted leads

Not pacing-dependent Device evaluation pre- and post-MR scan and with comparison of patients with recently implanted (< 42 days) leads

8 patients with recently implanted leads (7-36 days)

No complications in either the early or late group and no difference in parameters

One patient imaged 79 days after implant had frequent PVCs during scan with no action needed

Overall, statistically but not clinically significant changes seen after MR scan in R-wave voltage, ventricular threshold, and atrial impedance

MR imaging feasible in patients with recently implanted PMs

No clinically significant changes in function or on follow-up (average 104 days post-MRI)

Regression analysis of all 171 patients did not predict any change in pacing variables according to implant duration at time of scan

Regression analyses with generalized estimating equation models to compare pre- and post-MR scans, and to account for multiple scans in the same patient Small number of patients in the recently implanted group
Higgins et al22

Prospective, single-center cohort

N = 198 patients with 256 MR scans

Not pacing-dependent Incidence of POR in relation to device characteristics and patient characteristics

PORs occurred in 9 MRI scans in 8 patients and more frequently in Medtronic devices (P = .005) and devices released before 2002

POR caused decrease in heart rate (n = 4) and transient anomalous battery life indication in 1

POR infrequent and occurred in older generators (released prior to 2002) Pearson chi-squared for categorical variables, Wilcoxon rank sum or t test for continuous variables

Retrospective analysis of a small number of events and majority of patients in the entire database had Medtronic devices

Pacing-dependent patients excluded, for which clinical effects of a POR could have been more important

Naehle et al23

Prospective, single-center cohort

N = 18 patients, with 18 MR scans


Not pacing-dependent

At least 3 months from implantation


Lead parameters pre-, post- and at 3 months after MR scan

Serum troponin 1 hour before and 12 hours after MR scan

No significant changes in pacing threshold, impedance seen

No significant change in troponin observed

Battery voltage decreased from pre- to post-MR (P = .042)

In 2 scans oversensing as VF occurred but no attempt at therapy delivery was made

Troponin levels compared with Student t test, other comparisons with a Wilcoxon signed rank test Small case series
Higgins et al24

Retrospective, single-center cohort

N = 19 patients with abandoned leads (no generator) with N = 35 MR scans

Abandoned leads (no CIED generator)

Not pacing-dependent


Lead parameters

Mean of 1.63 abandoned leads per patient.

3 ICD leads, with 2 being dual coil

9 patients had long-term follow-up with no negative sequelae

No adverse events within 7 days of scan

When generator reimplanted (12 of 19 patients) there were no lead malfunctions or clinically significant changes in pacing threshold, but one patient had ventricular lead threshold that rose from 1.9 V to 2.6 V at 0.5 ms

Not provided

Small single center, retrospective

Unknown whether presence of a generator with functional leads could have affected results

No cardiac biomarkers analyzed

Strom et al42

Prospective, single-center cohort

N = 189 scans on 123 patients

Any PM or ICD; abandoned leads a relative contraindication


Clinical Utility

No deaths or system revision acutely or at 6 months

1 power on reset event

98.4% of scans were interpretable

75% of scans met pre-specified criteria for clinical utility

Event rates evaluated with binomial distribution

Linear mixed-effect models to evaluate system parameter changes

Kappa for evaluation of two-reader adjudication of utility

Strom et al


CIED =cardiac implantable electronic device; ICD =implantable cardioverter defibrillator; VF =ventricular fibrillation; POR =power-on reset; CRT=cardiac resynchronization therapy; SAR =specific absorption rate; DFT 5 defibrillation threshold test.
*e.g., mortality or morbidity %
†e.g., P value, hazard ratio, odds ratio, confidence intervals