- Email: [email protected]
MRI Safety for Patients Implanted With the MRI Ready ICD System
JACC: CLINICAL ELECTROPHYSIOLOGY
VOL.
-, NO. -, 2019
ª 2019 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION PUBLISHED BY ELSEVIER
MRI Safety for Patients Implanted With the MRI Ready ICD System MRI Ready Study Results Saman Nazarian, MD, PHD,a Daniel J. Cantillon, MD,b Pamela K. Woodard, MD,c Theofanie Mela, MD,d Adam M. Cline, MD, MSC,e Adam S. Strickberger, MD,f on behalf of the MRI Ready Investigators
ABSTRACT OBJECTIVE A prospective, multicenter study was performed to assess the safety and efficacy of the Durata and Optisure HV leads and the Ellipse VR implantable cardioverter-defibrillator (ICD) in a 1.5-T magnetic resonance imaging (MRI) environment. The primary safety objective was >90% freedom from MRI scan-related complications. The primary efficacy objectives were absence of change in capture threshold and absence of decrease of sensing amplitude from pre-MRI examination to 1 month after MRI. BACKGROUND MRI scanning of patients has been shown to be safe in patients with magnetic resonance-conditional implantable cardioverter-defibrillators (ICD) systems. METHODS Patients with a previously implanted magnetic resonance-conditional system underwent a nondiagnostic MRI scan. After the scan, a questionnaire was given to investigators and patients who returned for 1-month follow-up examination. A subset of patients underwent ventricular tachyarrhythmia or ventricular fibrillation (VT/VF) induction testing after the MRI to evaluate defibrillation function. RESULTS There were 220 patients (81% male; 62.1 11.2 years of age) enrolled who received an MRI scans from 29 centers. All primary safety and efficacy endpoints were met (p < 0.0001). No significant detection delays were found in 34 patients who had VT/VF episodes after the MRI scan was performed. Most physicians reported easy and acceptable programming and ease of MRI scheduling. CONCLUSIONS The MRI Ready MRI-conditional ICD system is safe, and electrical performance was not affected in patients receiving a 1.5-T whole-body MRI scan. Investigators reported favorable collaboration between cardiologists and radiologists in the MRI Ready IDE clinical trial. (A Clinical Evaluation of the Durata or Optisure High Voltage Leads and Ellipse VR ICD Undergoing MRI, an IDE Study [MRI Ready IDE]; NCT02787291) (J Am Coll Cardiol EP 2019;-:-–-) © 2019 by the American College of Cardiology Foundation.
From the aCardiac Physiology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania; bHeart and Vascular Institute, Cleveland Clinic Foundation; cMallinckrodt Institute of Radiology, Washington University School of Medicine; d f
Cardiac Arrhythmia Service, Harvard Medical School; eCardiac Arrhythmias and Heart Failure, Abbott Laboratories; and the
Arrhythmia, Inova Medical Group, Virginia. Sponsor: This study was supported by St. Jude Medical (now Abbott Laboratories). Dr.
Nazarian is a consultant for Biosense Webster, CardioSolv, Siemens, and ImriCor; and has received research grant support from the U.S. National Institutes of Health/National Heart, Lung, Blood Institute. Dr. Cantillon is a consultant for Abbott Laboratories and Boston Scientific. Dr. Woodard has received research support from U.S. National Institutes of Health/National Heart, Lung, and Blood Institute. Dr. Mela consults for and receives honoraria from Medtronic, Biotronik, and Abbott Laboratories. Dr. Cline is an employee of Abbott Laboratories. Dr. Strickberger consults for and receives honoraria from Abbott Laboratories. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page. Manuscript received December 10, 2018; revised manuscript received May 9, 2019, accepted May 13, 2019.
ISSN 2405-500X/$36.00
https://doi.org/10.1016/j.jacep.2019.05.010
2
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
C
ABBREVIATIONS AND ACRONYMS ICD = implantable
ardiac magnetic resonance imaging
successfully perform MRI scans in patients with an
(MRI) provides noninvasive multi-
MR-conditional ICD system.
planar
3-dimensional
cardioverter-defibrillator
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
2-dimensional (3D)
(2D)
whole-body
and
imaging
METHODS
with unparalleled spatial and temporal reso-
MRI = magnetic resonance imaging
lution that allows functional and tissue char-
TRIAL DESIGN AND PATIENT SELECTION. The MRI
SAR = specific absorption rate
acterization without exposure to ionizing
Ready IDE study was a prospective, multicenter,
radiation or potentially nephrotoxic iodin-
nonrandomized, single-arm clinical trial evaluating
VF = ventricular fibrillation VT = ventricular tachyarrhythmia
ated contrast dye (1). For these reasons, MRI
the safety and efficacy of an implanted single-
imaging has become the imaging modality
chamber MRI Ready ICD system in patients who had
of choice for diagnosing and staging across
undergone a pre-defined nondiagnostic MRI Ready
a wide spectrum of diseases, which has rapidly
IDE study (NCT02787291). This trial was sponsored by
increased the number of MRI scans performed while
the manufacturer, St. Jude Medical (Sylmar, Califor-
elevating the need for MRI imaging-conditional
nia). Tenets of the Declaration of Helsinki were fol-
implantable cardiac-defibrillators (ICD) (2). In 2007,
lowed, as were all laws and regulations of the
an estimated 27.5 million MRI imaging procedures
participating countries. Institutional Review Board/
were performed in the United States, in 7,195 hospi-
Ethics Committee approval and informed patient
tals and nonhospital sites (3). Moreover, forecast
consent were obtained for all participating centers
models estimate that 53% to 64% of ICD-eligible pa-
and patients.
tients may require MRI imaging within 10 years (4).
Inclusion criteria required patients to undergo
MRI imaging-conditional pacemakers have been
transvenous pectoral implant for a class I or II
approved by the U.S. Food and Drug Administration
guidelines-based ICD indication for at least 60 days
since 2008. However, in comparison to pacemakers,
and to be willing to undergo an elective MRI scan
ICD systems have added complexity due to larger
without sedation (12). The study visits and testing
bodied leads containing shocking coils and larger
regimens are outlined in Figure 1. Patients known to
generators carrying capacitors and transformers to
have previously implanted MR-unsafe devices or
produce sufficient voltage for cardioversion and
materials were excluded. All trial data were collected
defibrillation (5,6). Like MR-conditional pacemakers,
by trained personnel, and data integrity was main-
MR-conditional ICDs include less ferromagnetic ma-
tained by the sponsor through data monitoring and
terials (7,8), removal of Reed switches (9), addition
source
of current filters, and the need to disable tachyar-
follow-up.
verification
throughout
enrollment
and
rhythmia detection by programming to mitigate concerns for device movement, tissue injury due to
ICD SYSTEM. After meeting the enrollment criteria
heating, induced current, electrical resets, and
and providing consent, patients were enrolled in the
impaired device functionality (10). Previous studies
trial. The MRI Ready IDE system is composed of the
have not evaluated satisfaction with the necessary
Ellipse VR single-chamber ICD (models CD1411,
programming changes and operational workflow. It
CD1311, and CD1377, St. Jude Medical) connected to 1
is of paramount importance
that training and
of 2 active fixation, single-shock ICD Durata (model
education be provided to users of systems labeled
7120Q or 7122Q, St. Jude Medical) or Optisure
MR-conditional so that medical personnel clearly
(model LDA220Q or LDA210Q, St. Jude Medical) leads
understand what to look for during an MRI scan
in 58- to 65-cm lengths.
performed on a patient with the system. Training
MRI programming (in MRI imaging mode) was
also should occur to optimize patient monitoring,
initiated using the Merlin Patient Care System (Merlin
departmental workflow, and programming of MR
PCS, St. Jude Medical) that engages MRI parameter
hardware to MR-conditional mode and back to
settings for use during MRI scanning. The Merlin PCS
standard settings (11).
system has an onscreen MRI checklist that must be
The MRI Ready IDE study prospectively examined
completed before allowing MRI mode is allowed to be
the safety and functionality of a previously implanted
programmed. Items on the checklist include verifi-
MRI Ready ICD system when a 1.5-T MRI scan was
cation that capture thresholds are stable at #2.5 V for
being performed. To the best of the present authors’
0.5 ms, leads have proper MRI labeling, and no
knowledge, this is the first study that evaluated the
additional non-MRI imaging-compatible hardware is
ease of programming and the collaborative process
present. The MRI mode default settings disable
between radiologists and cardiologists needed to
tachycardia detection and therapy and turn off
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
- 2019:-–-
Nazarian et al. Safety of the MR-Conditional MRI Ready ICD System
F I G U R E 1 Study Flow Chart
pacing, or for pacing-dependent patients, selects an
enabled during post-scan device interrogation. A
asynchronous pacing mode with a base rate of 85
questionnaire was completed by the investigators to
ppm, which is maintained throughout the scan. After
evaluate the logistics and ease of programming and to
the MRI scan was performed, the MRI mode was
determine if there were any delays in reprogramming
switched off, and the normal ICD settings were
after each MRI scan was collected.
3
4
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
F I G U R E 2 Patient Accountability and Analysis Population
MRI SCAN. Study scans were performed using GE
25 min. The 25-min period was divided into 4 iden-
Healthcare
Kingdom),
tical Turbo Spin Echo (TSE) or TrueFISP sequences of
Siemens (Erlangen, Germany), or Phillips (Eindhoven,
6:15-min duration performed sequentially. The spe-
the Netherlands) scanners with a static magnetic field
cific study scan protocol was to maximize the MR
strength of 1.5 T, with radiofrequency excitation us-
effects on the MR ICD system and not allow for
ing the body coil to transmit in the frequency range of
diagnosis in a controlled scanning environment. The
63.75 0.5 MHz; a maximum gradient slew rate of
protocol for the head scan was developed to maxi-
200 T/m/s per axis was used. The ICD system was
mize the gradient exposure for 5 min; therefore, a
exposed to 2 worst-case MRI scan scenarios in normal
diffusion-weighted single-shot echo-planar imaging
scanner operating mode, first, a scan of the thorax
sequence was used with a head SAR of #3.2 W/kg.
where the radiofrequency component of the scan was
Throughout the scan procedure, pulse oximetry, and
maximized for whole-body-specific absorption rate
electrocardiography monitoring were conducted by a
(SAR), and second, a scan of the head where the
study-trained electrophysiologist, a cardiologist, or a
gradient component of the scan was maximized. The
clinician trained in Advanced Cardiac Life Support.
protocol for the thoracic scan was developed to ach-
Verbal communication with the patient took place
ieve a whole-body SAR of 2.0 W/kg for approximately
throughout the scan, and any clinically significant
(Little
Chalfont,
United
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
Nazarian et al.
- 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
complaints were recorded. If a patient was unable to tolerate the scan, the procedure was stopped, and a
T A B L E 1 Baseline Demographics and Clinical Characteristics
Primary Analysis Cohort (n ¼ 198)
repeat scan was not required. PRIMARY SAFETY ENDPOINT ANALYSIS. The pri-
Total Patient Cohort (N ¼ 220)
Age, yrs
mary safety endpoint was defined as freedom from
Mean SD
62.5 11.3
62.1 11.2
MRI scan-related complications for the Ellipse VR ICD
Range (min, max)
(29.0, 85.0)
(29.0, 85.0)
involving Durata or Optisure (RV) lead from MRI scan to 1-month after MRI scan testing of >90%. A complication was defined as a serious adverse device
Sex, n Females Males
41/220 (18.6) 179/220 (81.4)
Race
effect that required an invasive intervention or that
Asian
led to death. Reviews for safety events and correla-
Black or African American
tion with devices or procedures were performed by an
38/198 (19.2) 160/198 (80.8)
White
2/198 (1.0)
3/220 (1.4)
19/198 (9.6)
22/220 (10.0)
177/198 (89.4)
195/220 (88.6)
430.6 357.3
434.0 364.7
System age, days*
independent physician review board.
Mean SD
PRIMARY EFFECTIVENESS ENDPOINT ANALYSIS. The
Median
effectiveness endpoint was defined as the proportion
Range (min, max)
of Durata or Optisure (RV) leads implanted with the
Cardiovascular history
Ellipse VR ICD with ventricular capture threshold increase of #0.5 V at 0.5 ms and the proportion of right ventricular sensing amplitude decrease #50%
310.3
292
(73, 310.3)
(73, 292)
Dominant type of Cardiomyopathy Nonischemic Ischemic None
71/198 (35.9)
79/220 (35.9)
116/198 (58.6)
130/220 (59.1)
11/198 (5.6)
11/220 (5.0)
from pre-MRI scan testing to 1 month after the MRI
Coronary Artery Disease
126/198 (63.6)
141/220 (64.1)
scan. Device measurements were recorded before the
Myocardial Infarction
94/198 (47.5)
108/220 (49.1)
MRI scan, immediately afterward, and 1 month after
Unstable Angina
16/198 (8.1)
18/220 (8.2)
the MRI scan.
Prior Cardiac Interventions CABG
44/198 (22.2)
49/220 (22.3)
VT/VF EPISODE ANALYSIS. A subgroup of patients
PTCA /Stents /Atherectomy
84/198 (42.4)
95/220 (43.2)
gave consent for induction testing to evaluate the
Ablation
12/198 (6.1)
12/220 (5.5)
proportion of VT/VF episodes with significant detection delay assessed from device-detected VT/VF
Values are mean SD or n (%). *System age is defined as the time from implant of the last device until enrollment.
episodes (induced or spontaneous) with stored elec-
CABG ¼ coronary artery bypass graft; PTCA ¼ percutaneous transluminal coronary angioplasty.
trograms. An independent ventricular arrhythmia event review committee evaluated any clinically significant detection delay, defined as $5 s in this study. STATISTICAL
ANALYSIS. All
primary
hypothesis
endpoints were tested at the 2.5% significance level. The null hypothesis was rejected if the 97.5% lower confidence bound for freedom from MRI scan-related complications(s) or probability of success of ventricular capture was >90% for the safety endpoint and the ventricular capture endpoints. The sensing amplitude changes probability of success was defined as 87% for the second primary efficacy endpoint. The
Hungary, Poland, Spain, the United Kingdom, and the United States. In this trial, 220 MRI scans were performed successfully. Endpoint analysis data were not included for 22 patients due to the reasons given in Figure 2. Baseline characteristics and cardiovascular history are presented in Table 1; the demographics of the enrolled patients are consistent with a group of patients with an ICD needing an MRI scan. The mean follow-up duration for the study population was 2.3 1.0 months.
97.5% lower confidence bound was calculated by us-
MRI SCAN. MRI scans were performed in 220 pa-
ing the exact (Clopper-Pearson) method for binomial
tients. The average and mean whole-body SAR were
probabilities. Continuous variables were reported as
1.9 W/kg (88.9%; 176 of 198 patients had a whole-body
mean SD. Sample sizes were calculated using the
SAR of $1.8 W/kg) during the pre-defined non-
exact binomial method in SAS version 9.4 software
diagnostic scan sequences. There were no instances
(Cary, North Carolina).
of sustained or symptomatic tachy- or bradyarrhythmia episodes caused by or related to the MRI
RESULTS
scan; additionally, no lapse or loss of patient monitoring was reported during any of the scans.
Enrollment for this study began on May 19, 2016, and
Of the total number of patients, 209 completed the
concluded on January 12, 2017. A total of 220 patients
pre-MRI scan visit, had a 1-month post-MRI scan visit,
were enrolled from 29 sites (23 U.S. and 6 Europe) in
and thus were evaluated. Four patients did not have
5
6
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
original settings. The RV capture threshold of 1.25 V at
T A B L E 2 Primary Endpoint Results*
0.5 ms and sensing amplitude of 12.0 mV did not
Endpoint†
Proportion
97.5% LCB
PG
One-Sided p Value
Primary safety endpoint was freedom from MRI scan-related complications related to the ICD and RV lead
100.0 (198/198)
98.2
90
<.0001
Primary effectiveness endpoint: the RV threshold change success was ventricular capture threshold increase of #0.5 V for 0.5 ms
100.0 (187/187)
98.0
90
<.0001
RV capture success was ventricular sensing amplitude decrease of #50%
100.0 (198/198)
change between the pre-MRI scan and the post-MRI scan; the pacing lead impedance was 410 U pre-scan and 380 U post-scan, and the high-voltage lead impedance was 51 U and 49 U, respectively, for the pre-scan and post-scan device interrogation measurements. The cause of death was reported by the
98.2
87
<.0001
coroner as “sudden cardiac death,” and the patient died at home. An autopsy was not performed, and no further details were available.
*See Central Illustration. †All endpoints were measured from MRI scan to 1-month post-MRI scan testing. Exact binomial testing was used for statistical testing. ICD ¼ implantable cardioverter-defibrillator; LCB ¼ lower confidence bounds; MRI ¼ magnetic resonance imaging; PG ¼ performance goal; RV ¼ right ventricle/ventricular.
PRIMARY EFFECTIVENESS ENDPOINT. Ventricular
lead capture remained constant in 187 patients with a pre-, post-, and 1-month MRI scan visit measurements; averages were maintained at 0.8 0.3 V for
eligible study leads (3 patients had 52-cm-length
0.5 ms. There were no changes reported for any of
leads, which was not the lead length included in this
these groups. One patient did not have their capture
study, and 1 patient had a 7170Q Durata RV lead,
threshold collected at the post-MRI scan visit but was
which was not a lead model included in this study)
collected at the 1-month-post-MRI visit. The propor-
and were therefore excluded from the primary safety
tion RV leads implanted with the ICD with capture
endpoint analysis. Among the remaining 205 pa-
threshold increase of #0.5 V at 0.5 ms from pre-MRI
tients, 3 patients withdrew before 1-month post MRI
scan to 1-month post MRI scan was estimated at
scan visit, and 4 patients missed the 1-month post-
100%, and the 97.5% LCB was 98.0%, which is greater
MRI
than the PG of 90%. Hence, the null hypothesis was
scan
visit.
Therefore,
198
patients
who
completed the 1-month post MRI scan were included
rejected at the 2.5% level of significance, and this
as the analysis cohort for primary safety endpoint.
primary efficacy endpoint was met.
PRIMARY SAFETY ENDPOINT. For the primary safety
study endpoint of freedom from MRI scan-related complications(s) from MRI scan to 1 month after the MRI scan in 198 patients. None experienced MRI scanrelated complications(s) for the Durata or Optisure RV lead from MRI scan to 1 month after the MRI scan. The freedom from MRI scan-related complications(s) for the Durata or Optisure RV lead from MRI scan to 1 month after the MRI scan was estimated at 100%, and the 97.5% lower confidence bound (LCB) was 98.2%, which is greater than the performance goal (PG) of 90%. Hence, the null hypothesis was rejected at the 2.5% level of significance, and the primary safety endpoint was met (Table 2, Central Illustration). One patient death occurred during the study period, within 30 days after the MRI scan. The cause of death was adjudicated by the clinical events committee and was classified as unknown as to whether the death was related to the device or to the study. The patient was a 67-year-old male with ischemic cardiomyopathy,
atrial
fibrillation,
dyslipidemia,
diabetes, and peripheral vascular disease whose ICD
Sensing amplitudes of the R-wave showed insignificant decreases in the 198 patients between the 3 study visits. The mean RV sensing at baseline averaged 11.3 1.6 mV at the pre-MRI visit, 11.3 1.7 mV at the post-MRI visit, and 11.2 1.7 mV at the 1-month visit. The proportion of RV leads implanted with ICD with sensing amplitude decrease of #50% from preMRI scan to 1 month post-MRI scan was estimated at 100%, and the 97.5% LCB was 98.2%, which is greater than the PG of 87%. Hence, the null hypothesis was rejected at the 2.5% level of significance, and the second primary efficacy endpoint for the study was met. Both primary efficacy endpoints are illustrated in Table 2, Central Illustration. Other secondary endpoint device measurements were collected during the study that had no changes outside of the normal clinical ranges. RV impedance values at pre-MRI averaged 476.4 87.4 U and averaged 471.0 89.7 U at 1 month. High-voltage lead impedances showed few differences from the preMRI scan visit and the 1-month post MRI study visit at 68.1 16.2 U and 67.8 16.1 U, respectively.
had been implanted 269 days before the MRI scan was
VF/VT
performed. No adverse events or symptoms were re-
detected VF events (induced) were collected by 19
EPISODE
ANALYSIS. Thirty-two
ported during or after the scan, and the device pro-
11 days after the MRI scan from 27 patients. Each
gramming was confirmed to have been set back to the
patient had at least 1 VF episode. Of the 32 VF
device-
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
Nazarian et al.
- 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
C ENTR AL I LL U STRA T I O N Primary Endpoint Results
Endpoint* Primary safety endpoint- freedom from MRI scan related complications related to the ICD and RV lead Primary effectiveness endpoint: RV threshold change success is ventricular capture threshold increase of ≤0.5 V at 0.5 ms RV capture success is ventricular sensing amplitude decrease of ≤50%
Proportion
97.5% LCB
PG
One Sided p-Value
100.0% (198/198)
98.2%
90%
< .0001
100.0% (187/187)
98.0%
90%
< .0001
100.0% (198/198)
98.2%
87%
< .0001
Nazarian, S. et al. J Am Coll Cardiol EP. 2019;-(-):-–-. *All endpoints were measured from MRI scan to 1-month post-MRI scan testing. Exact binomial testing was used for statistical testing. ICD ¼ implantable cardioverterdefibrillator; LCB ¼ lower confidence bounds; MRI ¼ magnetic resonance imaging; PG ¼ performance goal; RV ¼ right ventricular.
episodes adjudicated by the Ventricular Arrhythmia
DISCUSSION
Event Review Committee (VAC), none had clinically significant detection delay of $5 s. Nine device-
The principal finding of this study is that the MRI
detected VT events (induced or spontaneous) were
Ready ICD system is safe and maintains functionality
collected by 54 26 days after the MRI scan from 7
in a 1.5-T MRI imaging environment and is associated
patients. Each patient had at least 1 VT episode
with a favorable operational workflow for the needed
(induced or spontaneous). Of the 9 VT episodes
programming changes. These findings are consistent
adjudicated by the VAC, none had clinically signifi-
with previous human studies of ICD systems designed
cant detection delay ($5 s). Therefore, there was no
specifically for full-body 1.5-T MRI scans, demon-
impact of MRI scans on the device’s ability to detect
strating safety without adverse effects on electrical
VF, nor VT episodes as summarized in Table 3.
performance or efficacy after scans (13). Importantly, this prospective study showed safety of full-body
SCHEDULING AND LOGISTICS. All investigators were
1.5-T MRI in an MR-conditional ICD system after a
asked to complete a questionnaire after all 220 study
median of 10.2 months (310.3 days) following im-
MRI scans were performed (220 of 220 [100%]). The
plantation, thus mimicking typical clinical time
subjective responses were collected and are summa-
frames for imaging following implantation and
rized in Table 4. Scheduling the MRI scan with the
distinguishing this trial from previous trials that
radiology department was reported as presenting the
consisted of de novo implants undergoing MRI scans.
most difficulty in the responses, although 90.9% (200
In addition, the study met safety and performance
of 220) of cases found scheduling easy or acceptable.
goals in many centers in diverse geographies. There
In cases where cardiac staff reviewed the MRI user’s
were no scan-related symptomatic or sustained tachy-
manual, 91.4% (201 of 220) of the time, cardiac staff
or bradyarrhythmia episodes observed during the
felt the information in the MRI user’s manual was
scanning sequence, consisting of a thorax scan
easy or acceptable to understand. Transportation and
directly over the system’s components and a head
transfer of patients from cardiology to radiology was
scan sequence reaching an SAR mean of 1.9 W/kg,
rated as easy or acceptable in 97.3% (200 of 220) of
above the nondiagnostic goal of 1.5 W/kg SAR levels
cases.
in-
set forth in the study design, as shown in comparable
vestigators pertaining to the programming of the de-
studies (14). There were no immediate MRI scan-
vices for MRI scans were found to be easy or
related complications for the Durata or the Optisure
acceptable.
RV study leads within 1 month of the scan. Our study
All
responses
from
the
cardiologist
7
8
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
event did not appear to be related to the system or the
T A B L E 3 VF and VT Episode Analysis
scan; however, it was treated in the “worst-case”
Variable
Induced
Spontaneous
Total
32 (27)
0
32 (27)
% of episodes with significant detection delay ($5 s)*
0.0 (0/32)
0
0.0 (0/32)
% of patients with significant detection delay ($5 s)†
0.0 (0/27)
0
0.0 (0/27)
VF Episodes
analysis of the primary safety and efficacy endpoints as a failure. The overall mortality rate of 0.5% per
Number of episodes (number of patients) that were adjudicated as true VF
month for the study cohort is similar to that of prior studies. Research reports a 1-year all-cause mortality rate in ICD patients of approximately 7% (15), which translates to a monthly all-cause mortality rate in the
VT Episodes Number of episodes (number of patients) that were adjudicated as true VT
1 (1)
8 (6)
9 (7)
% of episodes with significant detection delay ($5 s)*
0.0 (0/1)
0.0 (0/8)
0.0 (0/9)
% of patients with significant detection delay ($5 s)†
0.0 (0/1)
0.0 (0/6)
0.0 (0/7)
ICD population of 0.6%. A monthly all-cause mortality rate of 0.5% in this study is therefore not unexpected in this population due to disease prognosis. This study also collected information about the
Values are n/N or % (n/N). *n ¼ number with significant detection delay. †n ¼ patients with signification detection delay. VF ¼ ventricular fibrillation; VT ¼ ventricular tachyarrhythmia.
logistics and use of MRI mode programming. It was required that the cardiologist investigators had collaboration with the site’s radiology teams before trial site initiation. It was found that system pro-
also demonstrated robust detection of VT/VF episodes
gramming before the scan was considered easy or
in a subgroup of patients with either induced or
acceptable by most of its users. In 3.2% (n ¼ 7 of 220)
spontaneous arrhythmia episodes, which included a
of the responses to the questionnaire, there were
larger number of episodes among a bigger patient
difficulties found in the scheduling of the MRI
cohort of patients (32 VF episodes; 9 VT episodes). The
appointment with radiology. This low proportion may
authors view the need for robust data for the preser-
be due to the qualification requirement that sites
vation of ventricular arrhythmia detection after MRI
have a good working relationship between the 2 de-
scan, given the theoretical possibility for tissue injury
partments and that both departments approve the
or heating to diminish R wave sensing properties.
protocol before enrollment initiation. All other re-
There was 1 death reported during the study
sponses in the survey showed that it was not difficult
period, occurring 23 days after the MRI scan. The
to prepare a patient and set the ICD system’s MRI mode throughout the study. As discussed in the 2017 Heart Rhythm Society consensus (16), it is important
T A B L E 4 Questionnaire Responses of MRI Logistics and Ease of Use*
Questionnaire Rating Factor
to have collaboration between the radiologist and the
Easy
Acceptable
Difficult
NA
Cardiac staff ease of scheduling the CMR appointment with radiology
153 (69.5)
47 (21.4)
7 (3.2)
13 (5.9)
Cardiac staff ease of transporting/transferring patient to radiology for study scan
172 (78.2)
42 (19.1)
3 (1.4)
3 (1.4)
Cardiac staff level of comfort with monitoring and potentially resuscitating patient
178 (80.9)
42 (19.1)
0
0
Cardiac staff opinion of clarity of information in the MRI user’s manual, if the manual was reviewed
152 (69.1)
49 (22.3)
4 (1.8)
15 (6.8)
Ease of locating the MRI settings on the programmer
172 (78.2)
Scheduling and Logistics Between Cardiology and Radiology
cardiologist to maintain patient safety. STUDY LIMITATIONS. This study did not evaluate
image quality; this was not evaluated in this study to ensure that the pre-defined SAR limits were reached. Patients were excluded from the study if they had poor R-wave sensing or poor capture threshold; therefore, it is unknown how MRI scans may affect the system’s functionality in this patient population. Battery longevity data were not collected in this trial,
39 (17.7)
2 (0.9)
7 (3.2)
although the largest study to date of nonconditional systems showed no long-term effects on battery
MRI Setting Performance Ease of use of verifying the MRI settings via the MRI checklist on the programmer
174 (79.1)
39 (17.7)
0
7 (3.2)
Ease of programming the MRI settings on the programmer
176 (80.0)
37 (16.8)
0
7 (3.2)
Cardiac staff opinion of clarity of information on the MRI summary report on the programmer printout
167 (75.9)
50 (22.7)
0
3 (1.4)
Ease of use of disabling the MRI settings mode in the programmer
176 (80.0)
37 (16.8)
0
7 (3.2)
Overall usability of the MRI settings on the Merlin programmer
109 (49.5)
45 (20.5)
0
66 (30.0)
longevity
(17).
The
study
questionnaires
were
answered subjectively by study investigators, more defined measurements may decrease the knowledge gap for scheduling and MRI programming interaction of its users.
CONCLUSIONS The MRI-ready non-de novo MR-conditional ICD
*N ¼ 220 scans. A subjective questionnaire was completed by each investigator after each MRI scan procedure was conducted. Total number of responses was 220. NA ¼ not applicable.
system is safe and does not appear to affect electrical performance in the whole-body 1.5-T MRI scans for the MRI Ready ICD system, with good
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
Nazarian et al.
- 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
collaboration
between
cardiology
and
radiology
teams. ACKNOWLEDGMENTS The authors thank the MRI
Ready IDE Investigators and staff for their support in the MRI Ready Clinical trial. ADDRESS
FOR
PERSPECTIVES COMPETENCY IN MEDICAL KNOWLEDGE: Whole-body MRI scans are safe for patients with long-standing implanted MR-conditional ICD systems with strong collaboration between
CORRESPONDENCE:
Dr.
Saman
Nazarian, Cardiac Electrophysiology, The University of Pennsylvania Perelman School of Medicine, 3400 Spruce Street, Founders Pavillion 9118, Philadelphia, Pennsylvania 19104. E-mail: saman.nazarian@uphs.
radiology and cardiology teams. TRANSLATIONAL OUTLOOK: It is important to have strong collaboration for successful and safe MRI scans in patients with MR-conditional ICD systems in the 1.5-T scan environment.
upenn.edu.
REFERENCES 1. Rogulin A, Schlitter J, Vahlhaus C, Lombardi M, Brugada J, Vardas P, et al. Magnetic resonance imaging in individuals with cardiovascular
7. Luechinger R, Duru F, Scheidegger MB, et al. Force and torque effects of a 1.5-Tesla MRI scanner on cardiac pacemakers and ICDs. Pace 2001;
implantable electronic devices. Europace 2008; 10:336–46.
24:199–205.
2. Mavrogeni SI, Poulos G, Kolovou G, Theodorakis G. Magnetic resonance imagingconditional devices: Luxury or real clinical need? Hellenic J Cardiol 2017;58:256–60. 3. Medical Information Division. 2007 MRI Market Summary Report. Prepared for St. Jude Medical June 2008. 4. Nazarian S, Reynolds MR, Ryan MP, et al. Utilization of radiologic diagnostic imaging in patients with implantable cardiac defibrillators. J Magn Reson Imaging 2016;43:115–27. 5. Brugada J, Herse B, Sandsted B, et al. Clinical evaluation of defibrillation efficacy with a new single-capacitor biphasic waveform in patients undergoing implantation of an implantable cardioverter-defibrillator. Europace 2001;3:278–84. 6. Rinaldi CA, Simon RD, Geelen P, et al. A randomized prospective study of single coil versus dual coil defibrillation in patients with ventricular arrhythmias undergoing cardioverter defibrillator therapy. Pace 2003;26:1684–90.
8. Luechinger R, Zeijlemaker VA, Pedersen EM, et al. In vivo heating of pacemaker leads during magnetic resonance imaging. Eur Heart J 2005; 26:376–83. 9. Luechinger R, Duru F, Zeijemaker VA, et al. Pacemaker reed switch behavior in 0.5, 1.5, and 3.0 Tesla magnetic resonance imaging units: are reed switches always closed in strong magnetic fields? Pace 2002;25:1419–23. 10. Ainslie M, Miller C, Brown B, et al. Cardiac MRI of patients with implanted electrical cardiac devices. Heart 2014;100:363–9. 11. Shinbane JS, Colletti PM, Shellock FG. Magnetic resonance imaging in patients with cardiac pacemakers: era of “MR conditional” designs. J Cardiovasc Magn Reson 2011;13:63. 12. Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for Device-Based Therapy. J Am Coll Cardiol 2008;51:36–7.
randomized study. J AM Coll Cardiol 2015;65: 2581–8. 14. Awad K, Griffin J, Crawford TC, et al. Clinical safety of the Iforia implantable cardioverterdefibrillator system in patients subjected to thoracic spine and cardiac 1.5-T magnetic resonance scanning conditions. Heart Rhythm 2015;12: 2155–61. 15. Kobe J, Eckardt L. Early mortality in implantable cardioverter-defibrillator patients: from randomized controlled trials to real life. Europace 2009;11:694–6. 16. Indik JH, Gimbel JR, Abe H, et al. 2017 HRS expert consensus statement on magnetic resonance imaging and radiation exposure in patients with cardiovascular implantable electronic devices. Heart Rhythm 2017;14:97–153. 17. Nazarian S, Hansford R, Rahsepar AA, et al. Safety of magnetic resonance imaging in patients with cardiac devices. N Engl J Med 2017;377: 2555–64.
13. Gold MR, Sommer T, Schwitter J, et al. Full-
KEY WORDS implantable cardioverterdefibrillator, ICD system, magnetic resonance
body MRI in patients with an implantable cardioverter-defibrillator: primary results of a
imaging, MRI Ready IDE Study, Post MRI induction testing
9
VOL.
-, NO. -, 2019
ª 2019 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION PUBLISHED BY ELSEVIER
MRI Safety for Patients Implanted With the MRI Ready ICD System MRI Ready Study Results Saman Nazarian, MD, PHD,a Daniel J. Cantillon, MD,b Pamela K. Woodard, MD,c Theofanie Mela, MD,d Adam M. Cline, MD, MSC,e Adam S. Strickberger, MD,f on behalf of the MRI Ready Investigators
ABSTRACT OBJECTIVE A prospective, multicenter study was performed to assess the safety and efficacy of the Durata and Optisure HV leads and the Ellipse VR implantable cardioverter-defibrillator (ICD) in a 1.5-T magnetic resonance imaging (MRI) environment. The primary safety objective was >90% freedom from MRI scan-related complications. The primary efficacy objectives were absence of change in capture threshold and absence of decrease of sensing amplitude from pre-MRI examination to 1 month after MRI. BACKGROUND MRI scanning of patients has been shown to be safe in patients with magnetic resonance-conditional implantable cardioverter-defibrillators (ICD) systems. METHODS Patients with a previously implanted magnetic resonance-conditional system underwent a nondiagnostic MRI scan. After the scan, a questionnaire was given to investigators and patients who returned for 1-month follow-up examination. A subset of patients underwent ventricular tachyarrhythmia or ventricular fibrillation (VT/VF) induction testing after the MRI to evaluate defibrillation function. RESULTS There were 220 patients (81% male; 62.1 11.2 years of age) enrolled who received an MRI scans from 29 centers. All primary safety and efficacy endpoints were met (p < 0.0001). No significant detection delays were found in 34 patients who had VT/VF episodes after the MRI scan was performed. Most physicians reported easy and acceptable programming and ease of MRI scheduling. CONCLUSIONS The MRI Ready MRI-conditional ICD system is safe, and electrical performance was not affected in patients receiving a 1.5-T whole-body MRI scan. Investigators reported favorable collaboration between cardiologists and radiologists in the MRI Ready IDE clinical trial. (A Clinical Evaluation of the Durata or Optisure High Voltage Leads and Ellipse VR ICD Undergoing MRI, an IDE Study [MRI Ready IDE]; NCT02787291) (J Am Coll Cardiol EP 2019;-:-–-) © 2019 by the American College of Cardiology Foundation.
From the aCardiac Physiology, University of Pennsylvania Perelman School of Medicine, Philadelphia, Pennsylvania; bHeart and Vascular Institute, Cleveland Clinic Foundation; cMallinckrodt Institute of Radiology, Washington University School of Medicine; d f
Cardiac Arrhythmia Service, Harvard Medical School; eCardiac Arrhythmias and Heart Failure, Abbott Laboratories; and the
Arrhythmia, Inova Medical Group, Virginia. Sponsor: This study was supported by St. Jude Medical (now Abbott Laboratories). Dr.
Nazarian is a consultant for Biosense Webster, CardioSolv, Siemens, and ImriCor; and has received research grant support from the U.S. National Institutes of Health/National Heart, Lung, Blood Institute. Dr. Cantillon is a consultant for Abbott Laboratories and Boston Scientific. Dr. Woodard has received research support from U.S. National Institutes of Health/National Heart, Lung, and Blood Institute. Dr. Mela consults for and receives honoraria from Medtronic, Biotronik, and Abbott Laboratories. Dr. Cline is an employee of Abbott Laboratories. Dr. Strickberger consults for and receives honoraria from Abbott Laboratories. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page. Manuscript received December 10, 2018; revised manuscript received May 9, 2019, accepted May 13, 2019.
ISSN 2405-500X/$36.00
https://doi.org/10.1016/j.jacep.2019.05.010
2
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
C
ABBREVIATIONS AND ACRONYMS ICD = implantable
ardiac magnetic resonance imaging
successfully perform MRI scans in patients with an
(MRI) provides noninvasive multi-
MR-conditional ICD system.
planar
3-dimensional
cardioverter-defibrillator
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
2-dimensional (3D)
(2D)
whole-body
and
imaging
METHODS
with unparalleled spatial and temporal reso-
MRI = magnetic resonance imaging
lution that allows functional and tissue char-
TRIAL DESIGN AND PATIENT SELECTION. The MRI
SAR = specific absorption rate
acterization without exposure to ionizing
Ready IDE study was a prospective, multicenter,
radiation or potentially nephrotoxic iodin-
nonrandomized, single-arm clinical trial evaluating
VF = ventricular fibrillation VT = ventricular tachyarrhythmia
ated contrast dye (1). For these reasons, MRI
the safety and efficacy of an implanted single-
imaging has become the imaging modality
chamber MRI Ready ICD system in patients who had
of choice for diagnosing and staging across
undergone a pre-defined nondiagnostic MRI Ready
a wide spectrum of diseases, which has rapidly
IDE study (NCT02787291). This trial was sponsored by
increased the number of MRI scans performed while
the manufacturer, St. Jude Medical (Sylmar, Califor-
elevating the need for MRI imaging-conditional
nia). Tenets of the Declaration of Helsinki were fol-
implantable cardiac-defibrillators (ICD) (2). In 2007,
lowed, as were all laws and regulations of the
an estimated 27.5 million MRI imaging procedures
participating countries. Institutional Review Board/
were performed in the United States, in 7,195 hospi-
Ethics Committee approval and informed patient
tals and nonhospital sites (3). Moreover, forecast
consent were obtained for all participating centers
models estimate that 53% to 64% of ICD-eligible pa-
and patients.
tients may require MRI imaging within 10 years (4).
Inclusion criteria required patients to undergo
MRI imaging-conditional pacemakers have been
transvenous pectoral implant for a class I or II
approved by the U.S. Food and Drug Administration
guidelines-based ICD indication for at least 60 days
since 2008. However, in comparison to pacemakers,
and to be willing to undergo an elective MRI scan
ICD systems have added complexity due to larger
without sedation (12). The study visits and testing
bodied leads containing shocking coils and larger
regimens are outlined in Figure 1. Patients known to
generators carrying capacitors and transformers to
have previously implanted MR-unsafe devices or
produce sufficient voltage for cardioversion and
materials were excluded. All trial data were collected
defibrillation (5,6). Like MR-conditional pacemakers,
by trained personnel, and data integrity was main-
MR-conditional ICDs include less ferromagnetic ma-
tained by the sponsor through data monitoring and
terials (7,8), removal of Reed switches (9), addition
source
of current filters, and the need to disable tachyar-
follow-up.
verification
throughout
enrollment
and
rhythmia detection by programming to mitigate concerns for device movement, tissue injury due to
ICD SYSTEM. After meeting the enrollment criteria
heating, induced current, electrical resets, and
and providing consent, patients were enrolled in the
impaired device functionality (10). Previous studies
trial. The MRI Ready IDE system is composed of the
have not evaluated satisfaction with the necessary
Ellipse VR single-chamber ICD (models CD1411,
programming changes and operational workflow. It
CD1311, and CD1377, St. Jude Medical) connected to 1
is of paramount importance
that training and
of 2 active fixation, single-shock ICD Durata (model
education be provided to users of systems labeled
7120Q or 7122Q, St. Jude Medical) or Optisure
MR-conditional so that medical personnel clearly
(model LDA220Q or LDA210Q, St. Jude Medical) leads
understand what to look for during an MRI scan
in 58- to 65-cm lengths.
performed on a patient with the system. Training
MRI programming (in MRI imaging mode) was
also should occur to optimize patient monitoring,
initiated using the Merlin Patient Care System (Merlin
departmental workflow, and programming of MR
PCS, St. Jude Medical) that engages MRI parameter
hardware to MR-conditional mode and back to
settings for use during MRI scanning. The Merlin PCS
standard settings (11).
system has an onscreen MRI checklist that must be
The MRI Ready IDE study prospectively examined
completed before allowing MRI mode is allowed to be
the safety and functionality of a previously implanted
programmed. Items on the checklist include verifi-
MRI Ready ICD system when a 1.5-T MRI scan was
cation that capture thresholds are stable at #2.5 V for
being performed. To the best of the present authors’
0.5 ms, leads have proper MRI labeling, and no
knowledge, this is the first study that evaluated the
additional non-MRI imaging-compatible hardware is
ease of programming and the collaborative process
present. The MRI mode default settings disable
between radiologists and cardiologists needed to
tachycardia detection and therapy and turn off
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
- 2019:-–-
Nazarian et al. Safety of the MR-Conditional MRI Ready ICD System
F I G U R E 1 Study Flow Chart
pacing, or for pacing-dependent patients, selects an
enabled during post-scan device interrogation. A
asynchronous pacing mode with a base rate of 85
questionnaire was completed by the investigators to
ppm, which is maintained throughout the scan. After
evaluate the logistics and ease of programming and to
the MRI scan was performed, the MRI mode was
determine if there were any delays in reprogramming
switched off, and the normal ICD settings were
after each MRI scan was collected.
3
4
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
F I G U R E 2 Patient Accountability and Analysis Population
MRI SCAN. Study scans were performed using GE
25 min. The 25-min period was divided into 4 iden-
Healthcare
Kingdom),
tical Turbo Spin Echo (TSE) or TrueFISP sequences of
Siemens (Erlangen, Germany), or Phillips (Eindhoven,
6:15-min duration performed sequentially. The spe-
the Netherlands) scanners with a static magnetic field
cific study scan protocol was to maximize the MR
strength of 1.5 T, with radiofrequency excitation us-
effects on the MR ICD system and not allow for
ing the body coil to transmit in the frequency range of
diagnosis in a controlled scanning environment. The
63.75 0.5 MHz; a maximum gradient slew rate of
protocol for the head scan was developed to maxi-
200 T/m/s per axis was used. The ICD system was
mize the gradient exposure for 5 min; therefore, a
exposed to 2 worst-case MRI scan scenarios in normal
diffusion-weighted single-shot echo-planar imaging
scanner operating mode, first, a scan of the thorax
sequence was used with a head SAR of #3.2 W/kg.
where the radiofrequency component of the scan was
Throughout the scan procedure, pulse oximetry, and
maximized for whole-body-specific absorption rate
electrocardiography monitoring were conducted by a
(SAR), and second, a scan of the head where the
study-trained electrophysiologist, a cardiologist, or a
gradient component of the scan was maximized. The
clinician trained in Advanced Cardiac Life Support.
protocol for the thoracic scan was developed to ach-
Verbal communication with the patient took place
ieve a whole-body SAR of 2.0 W/kg for approximately
throughout the scan, and any clinically significant
(Little
Chalfont,
United
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
Nazarian et al.
- 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
complaints were recorded. If a patient was unable to tolerate the scan, the procedure was stopped, and a
T A B L E 1 Baseline Demographics and Clinical Characteristics
Primary Analysis Cohort (n ¼ 198)
repeat scan was not required. PRIMARY SAFETY ENDPOINT ANALYSIS. The pri-
Total Patient Cohort (N ¼ 220)
Age, yrs
mary safety endpoint was defined as freedom from
Mean SD
62.5 11.3
62.1 11.2
MRI scan-related complications for the Ellipse VR ICD
Range (min, max)
(29.0, 85.0)
(29.0, 85.0)
involving Durata or Optisure (RV) lead from MRI scan to 1-month after MRI scan testing of >90%. A complication was defined as a serious adverse device
Sex, n Females Males
41/220 (18.6) 179/220 (81.4)
Race
effect that required an invasive intervention or that
Asian
led to death. Reviews for safety events and correla-
Black or African American
tion with devices or procedures were performed by an
38/198 (19.2) 160/198 (80.8)
White
2/198 (1.0)
3/220 (1.4)
19/198 (9.6)
22/220 (10.0)
177/198 (89.4)
195/220 (88.6)
430.6 357.3
434.0 364.7
System age, days*
independent physician review board.
Mean SD
PRIMARY EFFECTIVENESS ENDPOINT ANALYSIS. The
Median
effectiveness endpoint was defined as the proportion
Range (min, max)
of Durata or Optisure (RV) leads implanted with the
Cardiovascular history
Ellipse VR ICD with ventricular capture threshold increase of #0.5 V at 0.5 ms and the proportion of right ventricular sensing amplitude decrease #50%
310.3
292
(73, 310.3)
(73, 292)
Dominant type of Cardiomyopathy Nonischemic Ischemic None
71/198 (35.9)
79/220 (35.9)
116/198 (58.6)
130/220 (59.1)
11/198 (5.6)
11/220 (5.0)
from pre-MRI scan testing to 1 month after the MRI
Coronary Artery Disease
126/198 (63.6)
141/220 (64.1)
scan. Device measurements were recorded before the
Myocardial Infarction
94/198 (47.5)
108/220 (49.1)
MRI scan, immediately afterward, and 1 month after
Unstable Angina
16/198 (8.1)
18/220 (8.2)
the MRI scan.
Prior Cardiac Interventions CABG
44/198 (22.2)
49/220 (22.3)
VT/VF EPISODE ANALYSIS. A subgroup of patients
PTCA /Stents /Atherectomy
84/198 (42.4)
95/220 (43.2)
gave consent for induction testing to evaluate the
Ablation
12/198 (6.1)
12/220 (5.5)
proportion of VT/VF episodes with significant detection delay assessed from device-detected VT/VF
Values are mean SD or n (%). *System age is defined as the time from implant of the last device until enrollment.
episodes (induced or spontaneous) with stored elec-
CABG ¼ coronary artery bypass graft; PTCA ¼ percutaneous transluminal coronary angioplasty.
trograms. An independent ventricular arrhythmia event review committee evaluated any clinically significant detection delay, defined as $5 s in this study. STATISTICAL
ANALYSIS. All
primary
hypothesis
endpoints were tested at the 2.5% significance level. The null hypothesis was rejected if the 97.5% lower confidence bound for freedom from MRI scan-related complications(s) or probability of success of ventricular capture was >90% for the safety endpoint and the ventricular capture endpoints. The sensing amplitude changes probability of success was defined as 87% for the second primary efficacy endpoint. The
Hungary, Poland, Spain, the United Kingdom, and the United States. In this trial, 220 MRI scans were performed successfully. Endpoint analysis data were not included for 22 patients due to the reasons given in Figure 2. Baseline characteristics and cardiovascular history are presented in Table 1; the demographics of the enrolled patients are consistent with a group of patients with an ICD needing an MRI scan. The mean follow-up duration for the study population was 2.3 1.0 months.
97.5% lower confidence bound was calculated by us-
MRI SCAN. MRI scans were performed in 220 pa-
ing the exact (Clopper-Pearson) method for binomial
tients. The average and mean whole-body SAR were
probabilities. Continuous variables were reported as
1.9 W/kg (88.9%; 176 of 198 patients had a whole-body
mean SD. Sample sizes were calculated using the
SAR of $1.8 W/kg) during the pre-defined non-
exact binomial method in SAS version 9.4 software
diagnostic scan sequences. There were no instances
(Cary, North Carolina).
of sustained or symptomatic tachy- or bradyarrhythmia episodes caused by or related to the MRI
RESULTS
scan; additionally, no lapse or loss of patient monitoring was reported during any of the scans.
Enrollment for this study began on May 19, 2016, and
Of the total number of patients, 209 completed the
concluded on January 12, 2017. A total of 220 patients
pre-MRI scan visit, had a 1-month post-MRI scan visit,
were enrolled from 29 sites (23 U.S. and 6 Europe) in
and thus were evaluated. Four patients did not have
5
6
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
original settings. The RV capture threshold of 1.25 V at
T A B L E 2 Primary Endpoint Results*
0.5 ms and sensing amplitude of 12.0 mV did not
Endpoint†
Proportion
97.5% LCB
PG
One-Sided p Value
Primary safety endpoint was freedom from MRI scan-related complications related to the ICD and RV lead
100.0 (198/198)
98.2
90
<.0001
Primary effectiveness endpoint: the RV threshold change success was ventricular capture threshold increase of #0.5 V for 0.5 ms
100.0 (187/187)
98.0
90
<.0001
RV capture success was ventricular sensing amplitude decrease of #50%
100.0 (198/198)
change between the pre-MRI scan and the post-MRI scan; the pacing lead impedance was 410 U pre-scan and 380 U post-scan, and the high-voltage lead impedance was 51 U and 49 U, respectively, for the pre-scan and post-scan device interrogation measurements. The cause of death was reported by the
98.2
87
<.0001
coroner as “sudden cardiac death,” and the patient died at home. An autopsy was not performed, and no further details were available.
*See Central Illustration. †All endpoints were measured from MRI scan to 1-month post-MRI scan testing. Exact binomial testing was used for statistical testing. ICD ¼ implantable cardioverter-defibrillator; LCB ¼ lower confidence bounds; MRI ¼ magnetic resonance imaging; PG ¼ performance goal; RV ¼ right ventricle/ventricular.
PRIMARY EFFECTIVENESS ENDPOINT. Ventricular
lead capture remained constant in 187 patients with a pre-, post-, and 1-month MRI scan visit measurements; averages were maintained at 0.8 0.3 V for
eligible study leads (3 patients had 52-cm-length
0.5 ms. There were no changes reported for any of
leads, which was not the lead length included in this
these groups. One patient did not have their capture
study, and 1 patient had a 7170Q Durata RV lead,
threshold collected at the post-MRI scan visit but was
which was not a lead model included in this study)
collected at the 1-month-post-MRI visit. The propor-
and were therefore excluded from the primary safety
tion RV leads implanted with the ICD with capture
endpoint analysis. Among the remaining 205 pa-
threshold increase of #0.5 V at 0.5 ms from pre-MRI
tients, 3 patients withdrew before 1-month post MRI
scan to 1-month post MRI scan was estimated at
scan visit, and 4 patients missed the 1-month post-
100%, and the 97.5% LCB was 98.0%, which is greater
MRI
than the PG of 90%. Hence, the null hypothesis was
scan
visit.
Therefore,
198
patients
who
completed the 1-month post MRI scan were included
rejected at the 2.5% level of significance, and this
as the analysis cohort for primary safety endpoint.
primary efficacy endpoint was met.
PRIMARY SAFETY ENDPOINT. For the primary safety
study endpoint of freedom from MRI scan-related complications(s) from MRI scan to 1 month after the MRI scan in 198 patients. None experienced MRI scanrelated complications(s) for the Durata or Optisure RV lead from MRI scan to 1 month after the MRI scan. The freedom from MRI scan-related complications(s) for the Durata or Optisure RV lead from MRI scan to 1 month after the MRI scan was estimated at 100%, and the 97.5% lower confidence bound (LCB) was 98.2%, which is greater than the performance goal (PG) of 90%. Hence, the null hypothesis was rejected at the 2.5% level of significance, and the primary safety endpoint was met (Table 2, Central Illustration). One patient death occurred during the study period, within 30 days after the MRI scan. The cause of death was adjudicated by the clinical events committee and was classified as unknown as to whether the death was related to the device or to the study. The patient was a 67-year-old male with ischemic cardiomyopathy,
atrial
fibrillation,
dyslipidemia,
diabetes, and peripheral vascular disease whose ICD
Sensing amplitudes of the R-wave showed insignificant decreases in the 198 patients between the 3 study visits. The mean RV sensing at baseline averaged 11.3 1.6 mV at the pre-MRI visit, 11.3 1.7 mV at the post-MRI visit, and 11.2 1.7 mV at the 1-month visit. The proportion of RV leads implanted with ICD with sensing amplitude decrease of #50% from preMRI scan to 1 month post-MRI scan was estimated at 100%, and the 97.5% LCB was 98.2%, which is greater than the PG of 87%. Hence, the null hypothesis was rejected at the 2.5% level of significance, and the second primary efficacy endpoint for the study was met. Both primary efficacy endpoints are illustrated in Table 2, Central Illustration. Other secondary endpoint device measurements were collected during the study that had no changes outside of the normal clinical ranges. RV impedance values at pre-MRI averaged 476.4 87.4 U and averaged 471.0 89.7 U at 1 month. High-voltage lead impedances showed few differences from the preMRI scan visit and the 1-month post MRI study visit at 68.1 16.2 U and 67.8 16.1 U, respectively.
had been implanted 269 days before the MRI scan was
VF/VT
performed. No adverse events or symptoms were re-
detected VF events (induced) were collected by 19
EPISODE
ANALYSIS. Thirty-two
ported during or after the scan, and the device pro-
11 days after the MRI scan from 27 patients. Each
gramming was confirmed to have been set back to the
patient had at least 1 VF episode. Of the 32 VF
device-
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
Nazarian et al.
- 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
C ENTR AL I LL U STRA T I O N Primary Endpoint Results
Endpoint* Primary safety endpoint- freedom from MRI scan related complications related to the ICD and RV lead Primary effectiveness endpoint: RV threshold change success is ventricular capture threshold increase of ≤0.5 V at 0.5 ms RV capture success is ventricular sensing amplitude decrease of ≤50%
Proportion
97.5% LCB
PG
One Sided p-Value
100.0% (198/198)
98.2%
90%
< .0001
100.0% (187/187)
98.0%
90%
< .0001
100.0% (198/198)
98.2%
87%
< .0001
Nazarian, S. et al. J Am Coll Cardiol EP. 2019;-(-):-–-. *All endpoints were measured from MRI scan to 1-month post-MRI scan testing. Exact binomial testing was used for statistical testing. ICD ¼ implantable cardioverterdefibrillator; LCB ¼ lower confidence bounds; MRI ¼ magnetic resonance imaging; PG ¼ performance goal; RV ¼ right ventricular.
episodes adjudicated by the Ventricular Arrhythmia
DISCUSSION
Event Review Committee (VAC), none had clinically significant detection delay of $5 s. Nine device-
The principal finding of this study is that the MRI
detected VT events (induced or spontaneous) were
Ready ICD system is safe and maintains functionality
collected by 54 26 days after the MRI scan from 7
in a 1.5-T MRI imaging environment and is associated
patients. Each patient had at least 1 VT episode
with a favorable operational workflow for the needed
(induced or spontaneous). Of the 9 VT episodes
programming changes. These findings are consistent
adjudicated by the VAC, none had clinically signifi-
with previous human studies of ICD systems designed
cant detection delay ($5 s). Therefore, there was no
specifically for full-body 1.5-T MRI scans, demon-
impact of MRI scans on the device’s ability to detect
strating safety without adverse effects on electrical
VF, nor VT episodes as summarized in Table 3.
performance or efficacy after scans (13). Importantly, this prospective study showed safety of full-body
SCHEDULING AND LOGISTICS. All investigators were
1.5-T MRI in an MR-conditional ICD system after a
asked to complete a questionnaire after all 220 study
median of 10.2 months (310.3 days) following im-
MRI scans were performed (220 of 220 [100%]). The
plantation, thus mimicking typical clinical time
subjective responses were collected and are summa-
frames for imaging following implantation and
rized in Table 4. Scheduling the MRI scan with the
distinguishing this trial from previous trials that
radiology department was reported as presenting the
consisted of de novo implants undergoing MRI scans.
most difficulty in the responses, although 90.9% (200
In addition, the study met safety and performance
of 220) of cases found scheduling easy or acceptable.
goals in many centers in diverse geographies. There
In cases where cardiac staff reviewed the MRI user’s
were no scan-related symptomatic or sustained tachy-
manual, 91.4% (201 of 220) of the time, cardiac staff
or bradyarrhythmia episodes observed during the
felt the information in the MRI user’s manual was
scanning sequence, consisting of a thorax scan
easy or acceptable to understand. Transportation and
directly over the system’s components and a head
transfer of patients from cardiology to radiology was
scan sequence reaching an SAR mean of 1.9 W/kg,
rated as easy or acceptable in 97.3% (200 of 220) of
above the nondiagnostic goal of 1.5 W/kg SAR levels
cases.
in-
set forth in the study design, as shown in comparable
vestigators pertaining to the programming of the de-
studies (14). There were no immediate MRI scan-
vices for MRI scans were found to be easy or
related complications for the Durata or the Optisure
acceptable.
RV study leads within 1 month of the scan. Our study
All
responses
from
the
cardiologist
7
8
Nazarian et al.
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019 - 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
event did not appear to be related to the system or the
T A B L E 3 VF and VT Episode Analysis
scan; however, it was treated in the “worst-case”
Variable
Induced
Spontaneous
Total
32 (27)
0
32 (27)
% of episodes with significant detection delay ($5 s)*
0.0 (0/32)
0
0.0 (0/32)
% of patients with significant detection delay ($5 s)†
0.0 (0/27)
0
0.0 (0/27)
VF Episodes
analysis of the primary safety and efficacy endpoints as a failure. The overall mortality rate of 0.5% per
Number of episodes (number of patients) that were adjudicated as true VF
month for the study cohort is similar to that of prior studies. Research reports a 1-year all-cause mortality rate in ICD patients of approximately 7% (15), which translates to a monthly all-cause mortality rate in the
VT Episodes Number of episodes (number of patients) that were adjudicated as true VT
1 (1)
8 (6)
9 (7)
% of episodes with significant detection delay ($5 s)*
0.0 (0/1)
0.0 (0/8)
0.0 (0/9)
% of patients with significant detection delay ($5 s)†
0.0 (0/1)
0.0 (0/6)
0.0 (0/7)
ICD population of 0.6%. A monthly all-cause mortality rate of 0.5% in this study is therefore not unexpected in this population due to disease prognosis. This study also collected information about the
Values are n/N or % (n/N). *n ¼ number with significant detection delay. †n ¼ patients with signification detection delay. VF ¼ ventricular fibrillation; VT ¼ ventricular tachyarrhythmia.
logistics and use of MRI mode programming. It was required that the cardiologist investigators had collaboration with the site’s radiology teams before trial site initiation. It was found that system pro-
also demonstrated robust detection of VT/VF episodes
gramming before the scan was considered easy or
in a subgroup of patients with either induced or
acceptable by most of its users. In 3.2% (n ¼ 7 of 220)
spontaneous arrhythmia episodes, which included a
of the responses to the questionnaire, there were
larger number of episodes among a bigger patient
difficulties found in the scheduling of the MRI
cohort of patients (32 VF episodes; 9 VT episodes). The
appointment with radiology. This low proportion may
authors view the need for robust data for the preser-
be due to the qualification requirement that sites
vation of ventricular arrhythmia detection after MRI
have a good working relationship between the 2 de-
scan, given the theoretical possibility for tissue injury
partments and that both departments approve the
or heating to diminish R wave sensing properties.
protocol before enrollment initiation. All other re-
There was 1 death reported during the study
sponses in the survey showed that it was not difficult
period, occurring 23 days after the MRI scan. The
to prepare a patient and set the ICD system’s MRI mode throughout the study. As discussed in the 2017 Heart Rhythm Society consensus (16), it is important
T A B L E 4 Questionnaire Responses of MRI Logistics and Ease of Use*
Questionnaire Rating Factor
to have collaboration between the radiologist and the
Easy
Acceptable
Difficult
NA
Cardiac staff ease of scheduling the CMR appointment with radiology
153 (69.5)
47 (21.4)
7 (3.2)
13 (5.9)
Cardiac staff ease of transporting/transferring patient to radiology for study scan
172 (78.2)
42 (19.1)
3 (1.4)
3 (1.4)
Cardiac staff level of comfort with monitoring and potentially resuscitating patient
178 (80.9)
42 (19.1)
0
0
Cardiac staff opinion of clarity of information in the MRI user’s manual, if the manual was reviewed
152 (69.1)
49 (22.3)
4 (1.8)
15 (6.8)
Ease of locating the MRI settings on the programmer
172 (78.2)
Scheduling and Logistics Between Cardiology and Radiology
cardiologist to maintain patient safety. STUDY LIMITATIONS. This study did not evaluate
image quality; this was not evaluated in this study to ensure that the pre-defined SAR limits were reached. Patients were excluded from the study if they had poor R-wave sensing or poor capture threshold; therefore, it is unknown how MRI scans may affect the system’s functionality in this patient population. Battery longevity data were not collected in this trial,
39 (17.7)
2 (0.9)
7 (3.2)
although the largest study to date of nonconditional systems showed no long-term effects on battery
MRI Setting Performance Ease of use of verifying the MRI settings via the MRI checklist on the programmer
174 (79.1)
39 (17.7)
0
7 (3.2)
Ease of programming the MRI settings on the programmer
176 (80.0)
37 (16.8)
0
7 (3.2)
Cardiac staff opinion of clarity of information on the MRI summary report on the programmer printout
167 (75.9)
50 (22.7)
0
3 (1.4)
Ease of use of disabling the MRI settings mode in the programmer
176 (80.0)
37 (16.8)
0
7 (3.2)
Overall usability of the MRI settings on the Merlin programmer
109 (49.5)
45 (20.5)
0
66 (30.0)
longevity
(17).
The
study
questionnaires
were
answered subjectively by study investigators, more defined measurements may decrease the knowledge gap for scheduling and MRI programming interaction of its users.
CONCLUSIONS The MRI-ready non-de novo MR-conditional ICD
*N ¼ 220 scans. A subjective questionnaire was completed by each investigator after each MRI scan procedure was conducted. Total number of responses was 220. NA ¼ not applicable.
system is safe and does not appear to affect electrical performance in the whole-body 1.5-T MRI scans for the MRI Ready ICD system, with good
JACC: CLINICAL ELECTROPHYSIOLOGY VOL.
-, NO. -, 2019
Nazarian et al.
- 2019:-–-
Safety of the MR-Conditional MRI Ready ICD System
collaboration
between
cardiology
and
radiology
teams. ACKNOWLEDGMENTS The authors thank the MRI
Ready IDE Investigators and staff for their support in the MRI Ready Clinical trial. ADDRESS
FOR
PERSPECTIVES COMPETENCY IN MEDICAL KNOWLEDGE: Whole-body MRI scans are safe for patients with long-standing implanted MR-conditional ICD systems with strong collaboration between
CORRESPONDENCE:
Dr.
Saman
Nazarian, Cardiac Electrophysiology, The University of Pennsylvania Perelman School of Medicine, 3400 Spruce Street, Founders Pavillion 9118, Philadelphia, Pennsylvania 19104. E-mail: saman.nazarian@uphs.
radiology and cardiology teams. TRANSLATIONAL OUTLOOK: It is important to have strong collaboration for successful and safe MRI scans in patients with MR-conditional ICD systems in the 1.5-T scan environment.
upenn.edu.
REFERENCES 1. Rogulin A, Schlitter J, Vahlhaus C, Lombardi M, Brugada J, Vardas P, et al. Magnetic resonance imaging in individuals with cardiovascular
7. Luechinger R, Duru F, Scheidegger MB, et al. Force and torque effects of a 1.5-Tesla MRI scanner on cardiac pacemakers and ICDs. Pace 2001;
implantable electronic devices. Europace 2008; 10:336–46.
24:199–205.
2. Mavrogeni SI, Poulos G, Kolovou G, Theodorakis G. Magnetic resonance imagingconditional devices: Luxury or real clinical need? Hellenic J Cardiol 2017;58:256–60. 3. Medical Information Division. 2007 MRI Market Summary Report. Prepared for St. Jude Medical June 2008. 4. Nazarian S, Reynolds MR, Ryan MP, et al. Utilization of radiologic diagnostic imaging in patients with implantable cardiac defibrillators. J Magn Reson Imaging 2016;43:115–27. 5. Brugada J, Herse B, Sandsted B, et al. Clinical evaluation of defibrillation efficacy with a new single-capacitor biphasic waveform in patients undergoing implantation of an implantable cardioverter-defibrillator. Europace 2001;3:278–84. 6. Rinaldi CA, Simon RD, Geelen P, et al. A randomized prospective study of single coil versus dual coil defibrillation in patients with ventricular arrhythmias undergoing cardioverter defibrillator therapy. Pace 2003;26:1684–90.
8. Luechinger R, Zeijlemaker VA, Pedersen EM, et al. In vivo heating of pacemaker leads during magnetic resonance imaging. Eur Heart J 2005; 26:376–83. 9. Luechinger R, Duru F, Zeijemaker VA, et al. Pacemaker reed switch behavior in 0.5, 1.5, and 3.0 Tesla magnetic resonance imaging units: are reed switches always closed in strong magnetic fields? Pace 2002;25:1419–23. 10. Ainslie M, Miller C, Brown B, et al. Cardiac MRI of patients with implanted electrical cardiac devices. Heart 2014;100:363–9. 11. Shinbane JS, Colletti PM, Shellock FG. Magnetic resonance imaging in patients with cardiac pacemakers: era of “MR conditional” designs. J Cardiovasc Magn Reson 2011;13:63. 12. Epstein AE, DiMarco JP, Ellenbogen KA, et al. ACC/AHA/HRS 2008 guidelines for Device-Based Therapy. J Am Coll Cardiol 2008;51:36–7.
randomized study. J AM Coll Cardiol 2015;65: 2581–8. 14. Awad K, Griffin J, Crawford TC, et al. Clinical safety of the Iforia implantable cardioverterdefibrillator system in patients subjected to thoracic spine and cardiac 1.5-T magnetic resonance scanning conditions. Heart Rhythm 2015;12: 2155–61. 15. Kobe J, Eckardt L. Early mortality in implantable cardioverter-defibrillator patients: from randomized controlled trials to real life. Europace 2009;11:694–6. 16. Indik JH, Gimbel JR, Abe H, et al. 2017 HRS expert consensus statement on magnetic resonance imaging and radiation exposure in patients with cardiovascular implantable electronic devices. Heart Rhythm 2017;14:97–153. 17. Nazarian S, Hansford R, Rahsepar AA, et al. Safety of magnetic resonance imaging in patients with cardiac devices. N Engl J Med 2017;377: 2555–64.
13. Gold MR, Sommer T, Schwitter J, et al. Full-
KEY WORDS implantable cardioverterdefibrillator, ICD system, magnetic resonance
body MRI in patients with an implantable cardioverter-defibrillator: primary results of a
imaging, MRI Ready IDE Study, Post MRI induction testing
9