Clinical Safety of Brain Magnetic Resonance Imaging with ...mriquestions.com/uploads/3/4/5/7/34572113/dbs_review_complicati… · MRI parameters Use only 1.5-T horizontal-bore MRI

No responsibility is assumed by Elsevier, its licensors or associates for any injury and/or damage to persons

or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Because of rapid advances in the

medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.

Clinical Safety of Brain Magnetic Resonance Imaging with Implanted Deep Brain Stimulation Hardware: Large Case

Series and Review of the Literature

Ludvic Zrinzo, Fumiaki Yoshid, Marwan I. Hariz, John Thornton, Thomas Foltynie, Tarek A. Yousry, Patricia Limousin

Ele

ctro

nic

Dis

trib

utio

n O

nly

Clinical Safety of Brain Magnetic Resonance Imaging with Implanted Deep BrainStimulation Hardware: Large Case Series and Review of the LiteratureLudvic Zrinzo1,2, Fumiaki Yoshida1, Marwan I. Hariz1,3, John Thornton4, Thomas Foltynie1,

Tarek A. Yousry4, Patricia Limousin1

INTRODUCTION

Over 75,000 patients have undergone deepbrain stimulation (DBS) procedures worldwide(28). Several centers use routine postoperativemagnetic resonance imaging (MRI) to verifyelectrode lead location and screen for postoper-ative complications. MRI is also an importantdiagnostic tool that may be required if DBS pa-tients develop additional unrelated symptoms.Additionally, functional MRI (fMRI) is a pow-erful research tool that may shed lightonto the mechanism of action of DBS andlead to potentially important clinical dis-coveries. Safety of MR imaging in patientswith implanted DBS hardware is thereforean important consideration.

Because there are concerns regarding thesafety of MRI with DBS hardware in situ (18,41), alternatives to MRI should be consid-ered when possible. However, an image-guided and image-verified approach to DBSrequires accurate verification of lead loca-tion within the intended target. Investiga-tions other than postoperative MRI may notprovide the required degree of accuracy andprecision. Image fusion of postoperativecomputed tomographic (CT) images to pre-operative MRI scans is open to fusion er-rors; mean errors of between 1.2 and 1.7mm have been reported when using fusionalgorithms, and larger errors of over 3 mm

may be expected in individual patients (30).Postoperative stereotactic CT or MRI priorto removal of the stereotactic frame aretherefore preferable methods of lead local-ization (10, 32, 50). Unfortunately, the ana-tomical target cannot be directly visualizedon CT; this method therefore relies on theassumption that anatomical structures havenot moved relative to the final lead locationwithin stereotactic space. MRI verificationallows both electrode artefact and intendedtarget to be visualized on the same image,and is therefore considered the gold stan-dard in verification of lead location in theclinical setting (17, 33, 45, 53, 54). Postop-

� BACKGROUND: Over 75,000 patients have undergone deep brain stimulation(DBS) procedures worldwide. Magnetic resonance imaging (MRI) is an importantclinical and research tool in analyzing electrode location, documenting postop-erative complications, and investigating novel symptoms in DBS patients.Functional MRI may shed light on the mechanism of action of DBS. MRI safety inDBS patients is therefore an important consideration.

� METHODS: We report our experience with MRI in patients with implantedDBS hardware and examine the literature for clinical reports on MRI safety withimplanted DBS hardware.

� RESULTS: A total of 262 MRI examinations were performed in 223 patientswith intracranial DBS hardware, including 45 in patients with an implanted pulsegenerator. Only 1 temporary adverse event occurred related to patient agitationand movement during immediate postoperative MR imaging. Agitation resolvedafter a few hours, and an MRI obtained before implanted pulse generatorimplantation revealed edema around both electrodes. Over 4000 MRI examina-tions in patients with implanted DBS hardware have been reported in theliterature. Only 4 led to adverse events, including 2 hardware failures, 1temporary and 1 permanent neurological deficit. Adverse neurological eventsoccurred in a unique set of circumstances where appropriate safety protocolswere not followed. MRI guidelines provided by DBS hardware manufacturers areinconsistent and vary among devices.

� CONCLUSIONS: The importance of MRI in modern medicine places pressureon industry to develop fully MRI-compatible DBS devices. Until then, theliterature suggests that, when observing certain precautions, cranial MR imagescan be obtained with an extremely low risk in patients with implanted DBShardware.

Key words� Deep brain stimulation� Magnetic resonance imaging� Safety

Abbreviations and AcronymsCT: Computed tomographyDBS: Deep brain stimulationfMRI: Functional magnetic resonance imagingIPG: Implanted pulse generatorMR: Magnetic resonanceMRI: Magnetic resonance imagingPD: Parkinson diseaseRF: RadiofrequencySAR: Specific absorption rateSTN: Subthalamic nucleus

From the 1Unit of FunctionalNeurosurgery, Sobell Department

of Motor Neuroscience and Movement Disorders, UCLInstitute of Neurology, University College London, QueenSquare, London, UK; 2Victor Horsley Department ofNeurosurgery, National Hospital for Neurology andNeurosurgery, Queen Square, London, UK; 3Departmentof Neurosurgery, University Hospital, Umeå, Sweden; and4UCL Institute of Neurology and Lysholm Department ofNeuroradiology, National Hospital for Neurology andNeurosurgery, Queen Square, London, UK.

To whom correspondence should be addressed:Ludvic Zrinzo, M.D., M.Sc. [E-mail: [email protected]]

Citation: World Neurosurg. (2011) 76, 1/2:164-172.DOI: 10.1016/j.wneu.2011.02.029

Journal homepage: www.WORLDNEUROSURGERY.org

Available online: www.sciencedirect.com

1878-8750/$ - see front matter © 2011 Elsevier Inc.All rights reserved.

PEER-REVIEW REPORTS

164 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2011.02.029

Ele

ctro

nic

Dis

trib

utio

n O

nly

erative MRI also provides the most compre-hensive assessment of radiological compli-cations after DBS because it is very sensitiveto postoperative edema and hemorrhage.

The interactions between the electro-magnetic fields of MRI machines and im-planted DBS hardware are complex andhave the potential to result in severe neuro-logical disability or death. The static mag-netic field, gradient magnetic field, and ra-diofrequency (RF) fields used for MRI mayinduce movement, heating, and electricalcurrent within the implanted cables andelectrodes or interfere with neurostimula-tor function (27). Phantom studies providesome insight into these effects but cannotreplicate every potential interaction in theclinical setting (11, 23, 37, 47).

After isolated reported complications re-lated to MRI, one DBS hardware manufac-turer changed its recommended safety pa-rameters (18, 41, 46). Current manufacturerguidelines are summarized in Table 1 (27,39, 40). Nevertheless, large numbers of MRexaminations in patients with implantedDBS hardware have been performed with-out complications; many of these MRIscans were reportedly obtained with pa-rameters outside current manufacturerguidelines.

AimsWe assess the safety of using MRI in pa-tients with implanted DBS hardware in alarge series of patients and examine the lit-erature for clinical reports on safety of MRIwith implanted DBS hardware.

METHODS

Our Case Series of MRI Performed withDBS In SituA prospective log of MRI scans performedin patients with implanted DBS hardware iskept at our institution. All complicationspotentially related to MR imaging are re-corded, as are the imaging parameters. Re-cords kept between the inauguration of theUnit of Functional Neurosurgery in Novem-ber 2002 and June 2010 inclusively were an-alyzed. We defined an MRI examination as asingle patient visit to the MRI scanner inwhich at least 1 acquisition was completed.

Management of Patients Before MR Imag-ing. Our image-guided approach to DBS

has been reported previously (12, 17, 20,33). Inability to tolerate surgery under localanesthesia and continual movement ofsome patients requiring surgery for move-ment disorders, coupled with increasing re-liance on verification of DBS lead locationon postoperative stereotactic MRI, has ledus to perform increasing numbers of DBSimplant procedures under general anesthe-sia. After DBS lead insertion, 28-cm-longDBS leads were either capped or connectedto a commercially available externalizationkit (3550-05, Medtronic, Minneapolis,Minnesota, USA). The caps or connectorswere tunneled to the parietal region (usuallyleft), and the excess DBS lead was coiledaround the relevant burr hole for between 1and 3 complete turns. Externalization wasperformed via separate stab incisions; thenoninsulated externalized wires were keptin isolation from each other and from theframe and secured to the forehead in loopswith adhesive tape.

Implanted pulse generators (IPGs) wereturned off prior to MR imaging, voltage setto zero, and bipolar settings applied ac-cording to default factory settings. Themagnetic reed switch was disabled in thosedevices where this function was available.

MRI Parameters Used. MR images were ob-tained from a single 1.5-T Signa MRI scan-ner (software version Lx 9.1, General Elec-tric, Milwaukee, Wisconsin, USA) using ahead transmit/receive coil that did not ex-tend over the chest to avoid RF exposure toareas outside the head and hence to mini-mize the area of the DBS circuit exposed.The MRI acquisition protocols were specif-ically designed to prevent the scanner-pre-dicted head-average specific absorptionrate (SAR) exceeding 0.4 W/kg while stillproviding adequate visualization of the tar-get anatomy. The gradient slew rate (dB/dt)was less than 20 T/s for all acquisitions. TheMRI imaging parameters used are pre-sented in Table 2. The electrode artefactwas noted on postoperative images, whichwere also examined for radiological com-plications.

Literature SearchAn online PubMed search of the Englishlanguage literature was performed on July 1,2010, using the search parameters: (“MRI”or “magnetic resonance imaging”) and(“safety” or “adverse event”) and (“deep

brain stimulation” or “DBS”). The titles andabstracts of reports were reviewed, and rel-evant articles were selected. Further articleswere selected based on quoted references.Publications on the safety of MRI with inva-sive EEG hardware were not considered fur-ther. Only articles that specifically com-mented on complications related to MRimaging or on the safety of MRI with im-planted DBS hardware were selected for fur-ther analysis. When relevant publicationsfrom the same institution had been pub-lished, only the article with the most detailsrelevant to MR imaging with DBS hardwarein situ was selected.

RESULTS

Hardware, Anatomical Targets, and DBSIndications at Our InstitutionMRI examinations were performed in pa-tients who had 28-cm-long DBS leads im-planted (models 3389 or 3387, Medtronic),except for 3 patients undergoing framelessstereotactic surgery when 40-cm versions ofthese DBS leads were used bilaterally and 3examinations in patients who underwentimplantation of bilateral pallidal leads 25cm in length (model 6146, St Jude Medical,Plano, Texas, USA). DBS targets includedthe subthalamic nucleus, posteroventralmotor and anteromedial limbic pallidum,ventrolateral motor and sensory thalamus,caudal zona incerta, periventricular/periaq-ueductal grey, posteromedial hypothala-mus, and pedunculopontine nucleus. Indi-cations for DBS included Parkinsondisease (PD), dystonia, tremor of variousnon-PD and nondystonic origins (essen-tial, posttraumatic, multiple sclerosis),gait freezing in progressive supranuclearpalsy, Gilles de la Tourette syndrome,chronic pain syndromes, and trigeminalautonomic cephalgias.

Experience with MRI and DBS Hardwarein Our InstitutionMRI Sessions Performed. A total of 262 MRIexaminations were performed in 223 patientswith intracranial DBS hardware during thestudy period. Of these, 45 MRI examinationswere performed in patients with intracranialDBS hardware connected to an IPG (Soletra orKinetra, Medtronic). Typical head averageSAR values for the gradient-echo localizer

PEER-REVIEW REPORTS

LUDVIC ZRINZO ET AL. MRI SAFETY WITH DBS HARDWARE

WORLD NEUROSURGERY 76 [1/2]: 164-172, JULY/AUGUST 2011 www.WORLDNEUROSURGERY.org 165

Ele

ctro

nic

Dis

trib

utio

n O

nly

ranged from 0.0039 to 0.0052 w/kg; for fastspin echo sequences, head average SAR wastypically 0.3603 to 0.3916 W/kg.

A total of 239 MRI examinations were

performed during 227 consecutive, frame-based, image-guided DBS procedures (141male; mean � SD age: 51.6 � 14.4 years).Only 2 of these procedures were not fol-

lowed by stereotactic MRI (technical rea-sons), and a nonstereotactic MRI was ob-tained by the first postoperative day. In 214examinations, DBS electrodes had been

Table 1. Current Manufacturer Guidelines Compared with Each Other and with the Protocol Observed in the Current Series

Medtronic St. Jude Medical Current Series

MRI should only be done ifabsolutely necessary, within thefollowing guidelines

MRI should only be done if absolutelynecessary, within the followingguidelines

Stereotactic MRI performedas routine after DBS leadimplantation

Hardwareconfiguration

– Implant system components asclose to the center line of thepatient as possible

–

Avoid implanting IPG in the abdomen Avoid implanting IPG in lower torso orabdomen

IPG occasionally implanted inabdomen

– Avoid separation of the extensionwhen implanting bilateralconfigurations

Minimal separation of extensionleads in bilateral implants

– DBS lead coiled �3 times aroundburr hole

DBS lead coiled 1 to 3 timesaround burr hole

Prior to performingMRI

Check neurostimulator functionbefore MRI examination

Check neurostimulator function beforeMRI examination

Neurostimulator functionchecked before MRIexamination

If possible, do not sedate the patientand ask to alert technician of anyunusual sensation

If possible, do not sedate the patientand ask to alert technician of anyunusual sensation

MRI performed in GA in�50% of patients

Externalisedsystems

Wrap the external portion withinsulating material

– No insulation material used

Keep the external portion out ofcontact with the patient

– Wires secured to patient’sskin with adhesive plaster

Keep the external leads straight withno loops and running down thecenter of the head coil

– Wires looped, kept separatefrom each other and frame

MRI parameters Use only 1.5-T horizontal-bore MRI Safety tests were only performed on1.5-T machines

1.5-T GE Signa scanner

Use only a transmit/receive head coil Use only transmit-receive head coil Transmit-receive head coil only

Limit the displayed average headSAR to 0.1 W/kg or less

Displayed local body SAR � 0.22W/kg (Brio); � 0.4 W/kg (Libra/Libra XP)

SAR � 0.4 W/kg

Limit the gradient dB/dT to 20 T/s orless

dB/dT limit � 20 T/s dB/dT � 20T/s

If an IPG ispresent

Interrogate IPG: if a broken lead issuspected (high impedance), do notperform an MRI

Interrogate IPG: if a broken lead issuspected (high impedance), do notperform an MRI

Interrogate IPG: if a broken leadis suspected (high impedance),do not perform an MRI

Turn stimulator output to off Turn stimulator output to off Turn stimulator output to off

Turn stimulation mode to bipolar – Turn stimulation mode tobipolar

Turn voltage amplitude to 0 Turn current amplitude to 0 Turn voltage amplitude to 0

Disable magnetic (reed) switch ifpresent

Disable IPG magnet (Brio) Disable magnetic (reed) switchif present

Discrepancies between manufacturer guidelines and between guidelines and our protocols have been highlighted in italic bold lettering. DBS, deep brain stimulation; GA, general anesthesia;IPG, implanted pulse generator; MRI, magnetic resonance imaging; SAR, specific absorption rate.

PEER-REVIEW REPORTS



Ele

ctro

nic

Dis

trib

utio

n O

nly

capped or externalized. DBS electrodeswere connected to a previous IPG in 25 MRIexaminations. One patient underwent a to-tal of 5 MRI sessions with implanted DBSelectrodes, 4 of which were with an im-planted IPG in situ over the course of afew years. Nine patients underwent 3 MRIscans with DBS hardware in situ, of which2 scans were with an implanted IPG.Three patients underwent 2 MRI scanswith hardware in situ, of which 1 scan waswith electrodes connected to an im-planted IPG. Two patients underwent 2MRI scans with brain electrodes not yetconnected to an IPG.

Frameless surgery was performed in 3patients who were unable to undergo sur-gery with the Leksell frame: the head wastoo large for frame placement in 1 patientand a pronounced thoracic gibbus pre-vented seating of the Leksell frame withinthe MRI cradle in the other two patients.Image-guided frameless DBS (Nexframe,Medtronic) with bilateral 40-cm DBSleads was therefore performed. Each pa-

tient underwent postoperative imagingwith DBS leads in situ.

The additional 20 MRI examinations (9 pa-tients, 3 male) were performed outside thecontext of the surgical procedure, in patientswith an implanted IPG connected to the DBSelectrodes, often to assess DBS electrode loca-tion in patients who had undergone surgeryoutside our program and sometimes to inves-tigate other clinical concerns.

Adverse Events in Our Patient Series Related toMRI. One patient became very dyskineticand agitated and his head moved during theimmediate postimplant stereotactic MRI,which had to be terminated before comple-tion of the scanning protocol. No abnor-mality was detected on the localizer se-quence, and agitation resolved after a fewhours. Nonstereotactic MRI was obtainedbefore IPG implantation on the secondpostoperative day and revealed a hyperin-tensity suggestive of edema around thewhole length of both electrodes (Figure 1).Head movement during the initial postop-

erative scan was thought to be the underly-ing cause. This patient had satisfactoryimprovement of PD symptoms with subtha-lamic nucleus (STN) DBS up to the last fol-low-up (5-years post surgery). There wereno other clinical or hardware-related ad-verse events associated with MRI acquisi-tion in any other patient.

Electrode Artefact Characteristics. On MRI,the electrode artefact is larger at the levelof the bare contacts than at the insulatedportion as reported previously (35, 52,54). The artefact is larger than the elec-trode dimensions, and is recognized as acentral signal void and surrounding hy-perintensity on all MRI sequences thatwere used (52).

It was noted that the size of the artefactcould vary between patients, and that inpatients with bilateral electrodes, theright artefact was always slightly largerthan the left, especially on proton-densityimages. The only exception was in thosepatients whose electrodes had been tun-

Table 2. MRI Protocols Used in the Current Series

Sequence T2W-FSE (STN) PDW-DSE (GPI) T1 volume

Coil Transmit-receive head coil Transmit-receive head coil Transmit-receive head coil

Scan plane Axial/coronal Axial/coronal Volumetric

Image mode 2-dimensional 2-dimensional 3-dimensional

Image options Flip angle: 90° Flip angle: 90° Flip angle: 30°

Scan timing TE � 95 ms TE � 15 ms TE � 2.3 ms

TR � 3000 ms TR � 4000 ms TR � 15.6 ms

Scan setup Receive bandwidth: 20.83 kHz Receive bandwidth: 15.63 kHz Receive bandwidth: 15.63 kHz

Scanning range Slice thickness: 2.0 mm, no gap Slice thickness: 2.0 mm, no gap Slice thickness: 2.0 mm, no gap

Slices: 8 Slices: 15 Slices: 124

Acquisition matrix (frequency): 256 (frequency): 256 (frequency): 256

(phase): 256 (phase): 256 (phase): 256

Phase field ofvies

1 1 1

250 mm 250 mm 250 mm

Number ofexcitations

4 4 (coronal), 3 (axial) 1

Contrast No No No

SAR �0.4 W/kg �0.4 W/kg �0.4 W/kg

Eco-train 9 7 N/A

Scan time 6:06 10:16 (coronal), 7:44 (axial) 5:21

DSE, dual spin echo; FSE, fast spin echo; GPI, globus pallidus internus; MRI, magnetic resonance imaging; N/A , not applicable; SAR, specific absorption rate; STN, subthalamic nucleus; TE,echo time; TR, time to repetition.

PEER-REVIEW REPORTS



Ele

ctro

nic

Dis

trib

utio

n O

nly

neled to the right parietal region, wherethe left artefact was larger than the right(Figure 2).

Literature ReviewTable 3 summarizes the clinical studies re-porting on the safety of MRI in patients withimplanted DBS hardware.

Case Reports of Adverse Events. There havebeen 4 reported adverse events while per-forming MRI scans in patients with im-planted DBS hardware (Table 3). In 1 pa-tient who underwent staged STN DBS, theends of the externalized leads were fixedoutside the transmit/receive head coil dur-ing an MRI scan of the brain at 1.0 T prior toIPG implantation (SAR not specified). Im-mediately after imaging, the patient devel-oped dystonic and ballistic movements of

the left leg that settled completely after afew weeks (41). Another patient had under-gone bilateral STN DBS for PD; the left elec-trode was connected to a Soletra IPG in theabdominal subcutaneous tissue and rightIPG placed in subclavian location. A lumbarMRI scan at 1.0 T with a full-body coil wasperformed 7 months after surgery (meanSAR � 1.26 W/kg) and caused thermocoag-ulation of brain tissue 2 to 3 cm around thetip of the left electrode, resulting in severepermanent neurological disability. The op-eration mode of the neurostimulator duringthe scan was not reported (17, 18). The onlyreported complications of MRI at 1.5 T(General Electric, Milwaukee, Wisconsin,USA) are 2 associated IPG failures withoutneurological sequelae (44, 49). No othercomplications of MRI with implanted DBShardware have been reported.

Safety of Structural MR in Patients with Im-planted DBS Hardware. Early reports of MRI

with implanted DBS hardware discussedsafety-related issues and reported no ad-verse effects in small numbers of patients(47, 48). Kovacs et al. (2006) publishedtheir experience with a 1.0-T MRI in 34DBS patients (24). They suggested thatresonant coupling of the RF pulses withimplanted hardware is more likely to leadto heating in a 1.0-T than a 1.5-T magnet,but they did not encounter adverse effectsin their series.

Larson et al. (25) in 2008 reported theirexperience obtaining 1071 MRI events at1.5 T in 405 patients with implanted DBShardware. No patients were scanned withexternalized leads. Both body transmit/head receive and head transmit-receivecoils were used, with SAR values well inexcess of 0.1 W/kg and of up to 3 W/kg,without adverse events.

In 2008, Vasques et al. (49) reported on226 MRI examinations under general an-esthesia without adverse neurological orradiological sequelae. However, they didreport 1 incidence of stimulator damage(Itrel III; Medtronic, Minneapolis, Min-nesota, USA) after spinal MRI. In addi-tion, the authors reported reset of stimu-lation parameters during cerebral MRI(Itrel III), a technical failure that is notinfrequent with this device outside thecontext of MRI acquisition (49).

Ondo et al. (2005) initially reportedthat postplacement MRI was routinelyperformed in 23 of 36 North Americancenters that responded to a survey (31). Ina further survey commissioned by the DBSStudy Group of the National ParkinsonFoundation, Tagliati et al. (44) in 2009reported on the safety of MRI in PD pa-tients with implanted DBS hardware. Atotal of 3304 PD patients with implantedhardware were reported to have under-gone MRI of the brain, and 177 DBS pa-tients had MRI of other regions. All 24centers that replied to the survey switchedthe stimulators off, and 23 of 24 set thevoltage to zero prior to imaging. The onlyadverse event reported was 1 IPG failureassociated with imaging in a 1.5-T scan-ner.

Chhabra et al. (9) report on 192 MRI ex-aminations during serial imaging of stagedbilateral implantation in 64 patients; eachpatient underwent 3 MRI examinationswith implanted hardware, the first after uni-lateral implantation without an implanted

Figure 1. (A) Axial slice from localizersequence of stereotactic MRI acquiredimmediately after electrode implantation.Large head movements occurred duringfurther scanning, and the patient becameagitated. The scan was terminated beforefull image acquisition. Agitation settledwithin a few hours. (B) NonstereotacticMRI scan was performed on the secondpostoperative day in the same patient.Oblique coronal reconstruction through leftelectrode in T2-weighted sequencerevealing significant edema along electrodetrack; at this point there was no furtheragitation. This patient continued to obtain asatisfactory improvement from STN DBSon PD symptoms 5 years after surgery.DBS, deep brain stimulation; MRI,magnetic resonance imaging; PD, Parkinsondisease; STN, subthalamic nucleus.

Figure 2. Axial proton-density stereotactic MRimages at the level of the foramen of Monroin 2 different patients with dystonia; DBSlead artifacts can be seen en route to theposteroventral internal pallidum. (A) Thecapped leads were tunneled to the leftparietal region (arrowhead), and the rightDBS lead artifact is larger. (B) The cappedleads were tunneled to the right parietalregion (arrowhead), and the left DBS leadartifact is larger. We hypothesize that thegreater number of coils around the leadipsilateral to the capped lead exerted agreater choking effect on induced current,resulting in a smaller MRI artifact (seeDiscussion). DBS, deep brain stimulation;MRI, magnetic resonance imaging.

PEER-REVIEW REPORTS



Ele

ctro

nic

Dis

trib

utio

n O

nly

IPG, the second during preoperative plan-ning of the second side with an implantedIPG, and the last after implantation of thesecond side. They suggest that subgalealcoils in the DBS electrode choke the flow ofinduced current, reducing potential heatingeffects. Although they report the safe use ofa receive-only head coil in this study on a1.5-T MRI machine, they comment thattheir institution now uses transmit-receivehead coils in view of the added margin ofsafety (9).

Fraix et al. (13) reported a large series of631 MRI examinations in patients withimplanted DBS hardware, including 61with IPG and 4 MR imaging of the spine,without adverse events. Voltage was set tozero, but other therapeutic parameterswere not adjusted prior to MRI. The reedswitch of the Kinetra was disabled and

that of the Itrel II set to so-called normalamplitude; this function was not availableon the Soletra. Nevertheless, they reportthat stimulation parameters remainedunchanged after the scan, regardless ofIPG used.

Nazzarro et al. (29) reported on 249 pa-tients undergoing 445 brain MRI scan ses-sions, including many implanted with anIPG without adverse events.

Weise et al. (51) reported their experienceof 243 MRI scans in 211 patients at 1.5 Tusing an RF transmit/receive coil with SARvalues of up to 0.9 W/kg, including imagingof externalized leads and leads connected toan implanted IPG, again without adverseeffects.

fMRI with Implanted DBS Electrodes. Rezai etal. (38) in 1999 reported the use of fMRI in

5 patients with implanted DBS electrodes.Jech et al. (21, 22) in 2001 reported per-forming fMRI studies in 4 PD patients.Hesselmann et al. (19) in 2004 reported asingle case of intraoperative fMRI. Aran-tes et al. (2) in 2006 also performed fMRIstudies in 4 patients. All of these studieswere performed on 1.5-T machines in pa-tients with unilateral implants that hadbeen externalized prior to permanent IPGimplantation, and stimulation was per-formed via an external impulse generator(Medtronic 3625) without adverse events.Stefurak et al. (43) in 2003 reported fMRIacquisition using a 1.5-T machine in apatient with bilateral externalized STNDBS electrodes, and Phillips et al. (34)reported the use of a 3.0-T machine toperform fMRI studies in 5 patients with

Table 3. Clinical Studies Reporting on the Safety of Magnetic Resonance Imaging in Patients with Implanted Deep BrainStimulation Hardware

Study

Number of MRIEvents/Events

with IPGNumber of

PatientsField

Strength (T) CoilSAR Value

(W/kg) dB/dt (T/s)ImplantedHardware

AdverseEvents Notes

Rezai et al., 1999 (38) 86/– 86 1.5 – – – Itrel II None –

Tronnier et al., 1999 (47) 25/20 20 0.25 and 1.5 Transmit-receive head,transmit-receive body

– – Itrel II and III None –

Uitti et al., 2002 (48) 11/11 5 1.5 – – – Itrel II None –

Spiegel et al., 2003 (41) Single casereport

1 1.0 Transmit-receive head – – Externalisedbilateral leads

Transienthemiballismus

Possible antennaeffect fromexternalised leads

Henderson et al., 2005 (18) Single casereport

1 1.0 Transmit-receive body 0.57–1.26 – Bilateral Soletra,one in abdomen

Permanentneurologicaldeficit

Operation mode ofSoletra unknown

Kovacs et al., 2006 (24) 34/34 34 1.0 Transmit-receive head �0.2 �19.87 T/s Soletra None –

Larson et al., 2008 (25) 1071/1071 405 1.5 Transmit-receive headand transmit body,receive head

�3.0 – Itrel II, Soletra,Kinetra, Libra

None –

Vasques et al., 2008 (49) 226/63 161 1.5 Transmit-receive headand transmit body,receive head

�1.9 – Itrel II and III,Soletra

One hardwarefailure

–

Tagliati et al., 2009 (44) – 3481 1.0 and 1.5 Various – – – One hardwarefailure

–

Chhabra et al., 2010 (9) 192/128 64 1.5 Receive only head �0.8 – Itrel II, Soletra None –

Fraix et al., 2010 (13) 631/61 570 1.0 and 1.5 Transmit body, receivehead

�4.0 – Itrel II, Soletra,Kinetra

None –

Nazzarro et al., 2010 (29) 445/– 249 1.0 and 1.5 Transmit-receive head 0.16–3.13 � 10.43T/s

Itrel II, Soletra None –

Weise et al., 2010 (51) 243/32 211 1.5 Transmit-receive head 0.8–0.9 �20 T/s Kinetra None –

This study 262/45 223 1.5 Transmit-receive head �0.4 � 20T/s Soletra, Kinetra One transientneurologicalevent

Patient movementduring MRI linkedto adverse event

DBS, deep brain stimulation; IPG, implanted pulse generator; MRI, magnetic resonance imaging; SAR, specific absorption rate.

PEER-REVIEW REPORTS



Ele

ctro

nic

Dis

trib

utio

n O

nly

externalized bilateral STN DBS electrodeswithout adverse events.

DISCUSSION

The present report adds a large series ofMRI examinations in patients with im-planted DBS hardware to the literature. Atotal of 223 patients underwent 262 MRIexaminations with average SAR values of�0.4 W/kg. Only 1 transient complicationoccurred, most probably related to largemovements of the head during scanning. Aliterature review revealed that MRI scanshave been obtained in over 4000 DBS pa-tients with only 4 complications related toMRI acquisition reported: 1 transient and 1permanent neurological complication, and2 IPG failures.

Interaction of DBS Hardware with MRIRF energy deposited in tissues during MRIcan cause tissue heating and is primarilydetermined by a single factor: the SAR (1).The U.S. Food and Drug Administrationspecifies an SAR threshold for head MRIexaminations of 3 W/kg averaged over thehead for any period of 10 minutes. However,these SAR safety thresholds may not applyin DBS patients. Hardware coupling to theRF field may result in substantial focal heat-ing that is influenced by the type and poweroutput of the transmit RF coil, as well as theshape, orientation, and type of implant(4-6, 11, 23, 36). Significant heating mayalso occur at breaks points in the implantedhardware, and MRI should be avoided if thisis suspected clinically (14). SAR values re-ported by commercial MRI scanners are ap-proximations based on simple geometrictissue models, with unspecified underlyingassumptions and margins of safety; scan-ner-reported SAR values may therefore varyamong different MRI systems (6), leadingsome groups to perform calorimetric cali-bration of head coil SAR estimates (15).

Interaction with strong magnetic fieldsmay result in movement of hardware, trau-matizing the surrounding brain (27). Al-though components are designed withminimally magnetic materials, head move-ment during an MRI study may amplify thiseffect due to Lorentz forces on conductingstructures. We hypothesize that this mech-anism resulted in the single event of tran-sient neurological symptoms in this study.

No further events have been noted uponcareful observation of subsequent DBS pa-tients during MR scanning. Notwithstand-ing variance from DBS manufacturer guide-lines (27, 39, 40), our group and others havenot observed any adverse events in DBS pa-tients undergoing MRI under general anes-thesia (26, 49).

Induced voltage within the lead systemmay also be induced by gradient magneticfields generated during an MRI scan and isdependent on the rate of change of mag-netic gradient pulses (dB/dT), loops in theDBS system, and location and orientation ofthe hardware with respect to the gradientcoils.

The physical arrangement of implantedhardware clearly has an impact on MRI in-teractions (37). Kainz et al. (23) point outthat increasing the number of extension ca-ble loops around the IPG may increase theheating effect dramatically. On the otherhand, phantom experiments have demon-strated that increasing the number of DBSlead loops around the burr hole signifi-cantly reduced induced temperature in-creases in a linear fashion (up to 50% in a1.5-T machine with 2.75 loops, P � 0.01)(4). Although the mechanism is unclear,loops that lie within the transmit-receivecoil are roughly perpendicular to the MRImagnetic field (B0), which may choke theflow of induced current (4, 9). In our study,we noticed that in bilaterally implanted pa-tients with a single dual channel IPG, thelead ipsilateral to the planned/actual IPGsite created a smaller artefact on MRI. Bothlead/cable connectors would be placed inthe parietal region ipsilateral to theplanned/actual IPG site. We hypothesizethat the greater number of coils around theburr hole in the ipsilateral lead exerted agreater choking effect on induced current,resulting in a smaller MRI artefact.

These findings have clear implicationson implantation techniques that may not bewidely recognized by functional neurosur-geons and are not uniformly mentioned incurrent manufacturer guidelines.

Clinical Experience with Implanted DBSHardwarePhantom studies have been very helpful inexploring the complex interaction betweenMRI and implanted hardware (3, 5-7, 11,23). It is clear that great care must be takenbefore performing such studies in clinical

practice. However, it is difficult to proposeuniversally applicable MRI safety guidelinesin view of the variability in techniques ofSAR calculation, the diversity of MRI sys-tems and software employed, and the in-creasing number of available DBS hardwaresystems and methods of implantation. Pub-lished recommendations from manufactur-ers have tried to take all of these factors intoaccount but are not consistent among man-ufacturers or devices, and have sometimessuggested parameters that severely restrictthe quality of acquired images or resultin significantly longer image acquisitiontimes (27).

Only the accumulation of clinical experi-ence will be able to provide a definitive mea-sure of MRI safety. Literature reports sug-gest that the risk of permanent neurologicalcomplications is around 1 in 4000. We con-sider accurate documentation of electrodelocation an essential part of the surgicalprocedure, and the risk from stereotacticMRI is certainly less than the risk of surgeryitself; it may even offer safety advantagesover neurophysiological methods of elec-trode confirmation that rely on brain map-ping with multiple brain penetrations (12).Both reports of adverse neurological out-comes occurred in the setting of a uniqueset of circumstances. The incidents in-volved 1.0-T machines, which use a differ-ent RF wavelength than the more com-monly available 1.5-T machines. Fieldinteractions between 1.0-T machines andDBS devices are less well characterized. Ahead transmit-receive coil was used in oneevent, a body transmit-receive coil in theother. SAR values were not reported in onestudy, and only estimated post hoc in theother. Externalized leads were straightenedand fixed outside the coil in one instance,and an abdominal IPG of unknown opera-tion mode was connected to the offendinglead in the other. It is imperative that if fur-ther adverse events do occur, they are inves-tigated thoroughly and comprehensively re-ported so that future similar events can beavoided.

Using a plastic or ceramic stylet duringsurgery would overcome the safety issues ofMRI localization with implanted hardware(32, 42). However, this approach involvesthe use of intraoperative MRI or the routinereopening of the cranial wounds for finalDBS electrode placement. In addition, a fur-ther MRI is required to provide radiological

PEER-REVIEW REPORTS



Ele

ctro

nic

Dis

trib

utio

n O

nly

documentation of final electrode contact lo-cation, and this approach does not addressthe issue of future MRI acquisition in DBSpatients.

MRI after DBS lead implantation pro-vides a sensitive assessment of potentialsurgical complications and is the gold stan-dard in anatomical verification of lead loca-tion, short of autopsy. Structural MRI is animportant diagnostic tool, and functionalMRI may also open new insights into themechanism of action of DBS. Notwith-standing manufacturer’s guidelines, phan-tom studies and a limited number of clinicalstudies suggest that fMRI may safely be per-formed during active stimulation (2, 8, 14,19, 21, 34, 38, 43). To our knowledge, fMRIwith implanted IPG has not yet been re-ported. However, a phantom study has sug-gested that with adequate safety controls,fully implanted systems may even offer asignificant safety advantage (8).

Physicians should be aware of the veryreal dangers of MRI in the presence of im-planted DBS hardware and liaise closelywith radiology and MR physicist colleaguesto establish local practices and monitoringof MRI events. They should also be aware ofthe reality that the number of reported ad-verse neurological events in such situationsis extremely low when sensible precautionsare taken, and many thousands of MRIscans have been obtained with implantedMRI hardware without adverse conse-quences.

Finally, manufacturers should be awareof the importance of MRI in modern medi-cine and strive to develop MRI-compatiblecomponents that would eliminate the risksin obtaining clinically relevant MR imagesin DBS patients (6, 16).

CONCLUSIONS

Together with other large reported series,experience at our institution confirms thatcranial MR images can be obtained in pa-tients with implanted DBS hardware, in-cluding with implanted IPGs, without seri-ous adverse events, when observing certainprecautions. This does not imply that thereare no safety concerns when performingMRI in patients with implanted DBS de-vices. The importance of MRI in modernmedicine places pressure on industry to de-velop fully MRI-compatible devices. Untilthen, the literature suggests that functional

neurosurgeons may be able to further min-imize potential risks of MRI acquisition inDBS patients by using longer leads that arethen coiled around the burr hole and avoid-ing excessively long extension cables thatwould require excessive coiling of redun-dant length around the IPG.

REFERENCES

1. Particular requirements for the safety of magneticresonance equipment for medical diagnosis. IEC60601-2-33, International Electrotechnical Com-mission, 2002.

2. Arantes PR, Cardoso EF, Barreiros MA, Teixeira MJ,Goncalves MR, Barbosa ER, Sukwinder SS, LeiteCC, Amaro E Jr: Performing functional magneticresonance imaging in patients with Parkinson’s dis-ease treated with deep brain stimulation. Mov Dis-ord 21:1154-1162, 2006.

3. Baker KB, Nyenhuis JA, Hrdlicka G, Rezai AR,Tkach JA, Shellock FG: Neurostimulation systems:assessment of magnetic field interactions associ-ated with 1.5- and 3-Tesla MR systems. J MagnReson Imaging 21:72-77, 2005.

4. Baker KB, Tkach J, Hall JD, Nyenhuis JA, ShellockFG, Rezai AR: Reduction of magnetic resonance im-aging-related heating in deep brain stimulationleads using a lead management device. Neurosur-gery 57:392-397; discussion 392-397, 2005.

5. Baker KB, Tkach J, Phillips MD: In vitro studies ofMRI-related heating of neurostimulation systems.Magn Reson Imaging 24:677-679; reply 679-680,2006.

6. Baker KB, Tkach JA, Nyenhuis JA, Phillips M, Shel-lock FG, Gonzalez-Martinez J, Rezai AR: Evaluationof specific absorption rate as a dosimeter of MRI-related implant heating. J Magn Reson Imaging 20:315-320, 2004.

7. Bhidayasiri R, Bronstein JM, Sinha S, Krahl SE, AhnS, Behnke EJ, Cohen MS, Frysinger R, Shellock FG:Bilateral neurostimulation systems used for deepbrain stimulation: in vitro study of MRI-relatedheating at 1.5 T and implications for clinical imag-ing of the brain. Magn Reson Imaging 23:549-555,2005.

8. Carmichael DW, Pinto S, Limousin-Dowsey P,Thobois S, Allen PJ, Lemieux L, Yousry T, ThorntonJS: Functional MRI with active, fully implanted,deep brain stimulation systems: safety and experi-mental confounds. Neuroimage 37:508-517, 2007.

9. Chhabra V, Sung E, Mewes K, Bakay RA, Abosch A,Gross RE: Safety of magnetic resonance imaging ofdeep brain stimulator systems: a serial imaging andclinical retrospective study. J Neurosurg 112:497-502, 2010.

10. Fiegele T, Feuchtner G, Sohm F, Bauer R, Anton JV,Gotwald T, Twerdy K, Eisner W: Accuracy of stereo-tactic electrode placement in deep brain stimulation

by intraoperative computed tomography.Parkinsonism Relat Disord 14:595-599, 2008.

11. Finelli DA, Rezai AR, Ruggieri PM, Tkach JA, Nyen-huis JA, Hrdlicka G, Sharan A, Gonzalez-Martinez J,Stypulkowski PH, Shellock FG: MR imaging-relatedheating of deep brain stimulation electrodes: in vitrostudy. AJNR Am J Neuroradiol 23:1795-1802, 2002.

12. Foltynie T, Zrinzo L, Martinez-Torres I, Tripoliti E,Petersen E, Holl E, Aviles-Olmos I, Jahanshahi M,Hariz M, Limousin P: MRI-guided STN DBS in Par-kinson’s disease without microelectrode recording:efficacy and safety. J Neurol Neurosurg Psychiatry82:358-363, 2011.

13. Fraix V, Chabardes S, Krainik A, Seigneuret E,Grand S, Le Bas JF, Krack P, Benabid AL, Pollak P:Effects of magnetic resonance imaging in patientswith implanted deep brain stimulation systems. JNeurosurg 113:1242-1245, 2010.

14. Georgi JC, Stippich C, Tronnier VM, Heiland S: Ac-tive deep brain stimulation during MRI: a feasibilitystudy. Magn Reson Med 51:380-388, 2004.

15. Gorny KR, Bernstein MA, Felmlee JP, Ward HA, Mc-Gee KP, Lanners DM, Lee KH: Calorimetric calibra-tion of head coil SAR estimates displayed on a clini-cal MR scanner. Phys Med Biol 53:2565-2576, 2008.

16. Greatbatch W, Miller V, Shellock FG: Magnetic res-onance safety testing of a newly developed fiber-optic cardiac pacing lead. J Magn Reson Imaging16:97-103, 2002.

17. Hariz MI, Krack P, Melvill R, Jorgensen JV, HamelW, Hirabayashi H, Lenders M, Wesslen N, TengvarM, Yousry TA: A quick and universal method forstereotactic visualization of the subthalamic nucleusbefore and after implantation of deep brain stimula-tion electrodes. Stereotact Funct Neurosurg 80:96-101, 2003.

18. Henderson JM, Tkach J, Phillips M, Baker K, Shell-ock FG, Rezai AR: Permanent neurological deficitrelated to magnetic resonance imaging in a patientwith implanted deep brain stimulation electrodesfor Parkinson’s disease: case report. Neurosurgery57:E1063; discussion E1063, 2005.

19. Hesselmann V, Sorger B, Girnus R, Lasek K, Maar-ouf M, Wedekind C, Bunke J, Schulte O, Krug B,Lackner K, Sturm V: Intraoperative functional MRIas a new approach to monitor deep brain stimula-tion in Parkinson’s disease. Eur Radiol 14:686-690,2004.

20. Holl EM, Petersen EA, Foltynie T, Martinez-Tor-res I, Limousin P, Hariz MI, Zrinzo L: Improvingtargeting in image-guided frame-based deepbrain stimulation. Neurosurgery 67(2 Suppl Op-erative):437-447, , 2010.

21. Jech R, Urgosik D, Tintera J, Nebuzelsky A, Krasen-sky J, Liscak R, Roth J, Ruzicka E: Functional mag-netic resonance imaging during deep brain stimula-tion: a pilot study in four patients with Parkinson’sdisease. Mov Disord 16:1126-1132, 2001.

22. Journee HL: Comment on Jech et al, fMRI imagingduring deep brain stimulation. Mov Disord 18:460-461; reply 461-462, 2003.

PEER-REVIEW REPORTS



email: [email protected] A81485

Ele

ctro

nic

Dis

trib

utio

n O

nly

23. Kainz W, Neubauer G, Uberbacher R, Alesch F,Chan DD: Temperature measurement on neurolog-ical pulse generators during MR scans. Biomed EngOnline 1:2, 2002.

24. Kovacs N, Nagy F, Kover F, Feldmann A, Llu-miguano C, Janszky J, Kotek G, Doczi T, Balas I:Implanted deep brain stimulator and 1.0-Tesla mag-netic resonance imaging. J Magn Reson Imaging24:1409-1412, 2006.

25. Larson PS, Richardson RM, Starr PA, Martin AJ:Magnetic resonance imaging of implanted deepbrain stimulators: experience in a large series. Ste-reotact Funct Neurosurg 86:92-100, 2008.

26. Maldonado IL, Roujeau T, Cif L, Gonzalez V, El-Fertit H, Vasques X, Bonafe A, Coubes P: Magneticresonance-based deep brain stimulation technique:a series of 478 consecutive implanted electrodeswith no perioperative intracerebral hemorrhage.Neurosurgery 65:196-201; discussion 201–192,2009.

27. Medtronic: MRI Guidelines for Medtronic DeepBrain Stimulation Systems: Physician’s Manual.Minneapolis: Medtronic; 2006:1–20.

28. Medtronic:DBSforParkinson’sDisease.Patient infor-mation brochure. Minneapolis: Medtronic; 2010:1-32.

29. Nazzaro JM, Lyons KE, Wetzel LH, Pahwa R: Use ofbrain MRI after deep brain stimulation hardwareimplantation. Int J Neurosci 120:176-183, 2010.

30. O’Gorman RL, Jarosz JM, Samuel M, Clough C, Sel-way RP, Ashkan K: CT/MR image fusion in the post-operative assessment of electrodes implanted fordeep brain stimulation. Stereotact Funct Neurosurg87:205-210, 2009.

31. Ondo WG, Bronte-Stewart H: The North Americansurvey of placement and adjustment strategies fordeep brain stimulation. Stereotact Funct Neurosurg83:142-147, 2005.

32. Patel NK, Plaha P, O’Sullivan K, McCarter R, Hey-wood P, Gill SS: MRI directed bilateral stimulationof the subthalamic nucleus in patients with Parkin-son’s disease. J Neurol Neurosurg Psychiatry 74:1631-1637, 2003.

33. Petersen EA, Holl EM, Martinez-Torres I, Foltynie T,Limousin P, Hariz MI, Zrinzo L: Minimizing brainshift in stereotactic functional neurosurgery. Neu-rosurgery 67:213-221; discussion 221, 2010.

34. Phillips MD, Baker KB, Lowe MJ, Tkach JA, CooperSE, Kopell BH, Rezai AR: Parkinson disease: patternof functional MR imaging activation during deepbrain stimulation of subthalamic nucleus—initialexperience. Radiology 239:209-216, 2006.

35. Pollo C, Villemure JG, Vingerhoets F, Ghika J,Maeder P, Meuli R: Magnetic resonance artifact in-duced by the electrode Activa 3389: an in vitro and in

vivo study. Acta Neurochir (Wien) 146:161-164,2004.

36. Rezai AR, Baker KB, Tkach JA, Phillips M, HrdlickaG, Sharan AD, Nyenhuis J, Ruggieri P, Shellock FG,Henderson J: Is magnetic resonance imaging safefor patients with neurostimulation systems used fordeep brain stimulation? Neurosurgery 57:1056-1062; discussion 1056-1062, 2005.

37. Rezai AR, Finelli D, Nyenhuis JA, Hrdlicka G, TkachJ, Sharan A, Rugieri P, Stypulkowski PH, ShellockFG: Neurostimulation systems for deep brain stim-ulation: in vitro evaluation of magnetic resonanceimaging-related heating at 1.5 Tesla. J Magn ResonImaging 15:241-250, 2002.

38. Rezai AR, Lozano AM, Crawley AP, Joy ML, DavisKD, Kwan CL, Dostrovsky JO, Tasker RR, MikulisDJ: Thalamic stimulation and functional magneticresonance imaging: localization of cortical and sub-cortical activation with implanted electrodes. Tech-nical note. J Neurosurg 90:583-590, 1999.

39. St Jude Medical Inc: Libra Deep Brain StimulationSystem: Clinician’s Manual. Plano, TX: St. JudeMedical Company; 2008.

40. St Jude Medical Inc: Brio Deep Brain StimulationSystem: Clinician’s Manual. Plano, TX: St. JudeMedical Company; 2009.

41. Spiegel J, Fuss G, Backens M, Reith W, Magnus T,Becker G, Moringlane JR, Dillmann U: Transientdystonia following magnetic resonance imaging in apatient with deep brain stimulation electrodes forthe treatment of Parkinson disease. Case report. JNeurosurg 99:772-774, 2003.

42. Starr PA, Martin AJ, Ostrem JL, Talke P, Levesque N,Larson PS: Subthalamic nucleus deep brain stimu-lator placement using high-field interventionalmagnetic resonance imaging and a skull-mountedaiming device: technique and application accuracy. JNeurosurg 112:479-490, 2010.

43. Stefurak T, Mikulis D, Mayberg H, Lang AE, Hev-enor S, Pahapill P, Saint-Cyr J, Lozano A: Deep brainstimulation for Parkinson’s disease dissociatesmood and motor circuits: a functional MRI casestudy. Mov Disord 18:1508-1516, 2003.

44. Tagliati M, Jankovic J, Pagan F, Susatia F, Isaias IU,Okun MS: Safety of MRI in patients with implanteddeep brain stimulation devices. Neuroimage47(suppl 2):T53-57, 2009.

45. Tisch S, Zrinzo L, Limousin P, Bhatia KP, Quinn N,Ashkan K, Hariz M: Effect of electrode contact loca-tion on clinical efficacy of pallidal deep brain stim-ulation in primary generalised dystonia. J NeurolNeurosurg Psychiatry 78:1314-1319, 2007.

46. Tremmel J: Urgent Device Correction—Change ofSafe Limits for MRI Procedures Used with theMedtronic Activa Deep Brain Stimulation Systems.Minneapolis: Medtronic; 2005.

47. Tronnier VM, Staubert A, Hahnel S, Sarem-Aslani A:Magnetic resonance imaging with implanted neu-rostimulators: an in vitro and in vivo study. Neuro-surgery 44:118-125; discussion 125–116, 1999.

48. Uitti RJ, Tsuboi Y, Pooley RA, Putzke JD, Turk MF,Wszolek ZK, Witte RJ, Wharen RE Jr: Magnetic res-onance imaging and deep brain stimulation. Neuro-surgery 51:1423-1428; discussion 1428-1431, 2002.

49. Vasques X, Tancu C, Cif L, Biolsi B, Maldonado I,Bonafe A, Bars E, Coubes P: Cerebral magnetic res-onance imaging feasibility in patients with im-planted neurostimulation system for deep brainstimulation. Open Magn Reson J 1:1-8, 2008.

50. Vayssiere N, Hemm S, Zanca M, Picot MC, Bonafe A,Cif L, Frerebeau P, Coubes P: Magnetic resonanceimaging stereotactic target localization for deepbrain stimulation in dystonic children. J Neurosurg93:784-790, 2000.

51. Weise LM, Schneider GH, Kupsch A, Haumesser J,Hoffmann KT: Postoperative MRI examinations inpatients treated by deep brain stimulation using anon-standard protocol. Acta Neurochir (Wien) 152:2021-2027, 2010.

52. Yelnik J, Damier P, Demeret S, Gervais D, BardinetE, Bejjani BP, Francois C, Houeto JL, Arnule I, Dor-mont D, Galanaud D, Pidoux B, Cornu P, Agid Y:Localization of stimulating electrodes in patientswith Parkinson disease by using a three-dimen-sional atlas-magnetic resonance imaging coregis-tration method. J Neurosurg 99:89-99, 2003.

53. Zrinzo L: The role of imaging in the surgical treat-ment of movement disorders. Neuroimaging ClinNorth Am 20:125-140, 2010.

54. Zrinzo L, van Hulzen AL, Gorgulho AA, Limousin P,Staal MJ, De Salles AA, Hariz MI: Avoiding the ven-tricle: a simple step to improve accuracy of anatom-ical targeting during deep brain stimulation. J Neu-rosurg 110:1283-1290, 2009.

Conflict of interest statement: This work was undertakenat UCL/UCLH and was funded in part by the Department ofHealth NIHR Biomedical Research Centres funding scheme.The Unit of Functional Neurosurgery, UCL Institute ofNeurology, Queen Square, London, is supported by theParkinson’s Appeal. Dr. Zrinzo, Dr. Hariz, Dr. Foltynie, andDr. Limousin occasionally receive honoraria and travelcosts for educational activities from the manufacturer ofthe devices used in this study. The authors have nopersonal financial or institutional interest in any of thedrugs, materials, or devices described in this article.

received 20 December 2010; accepted 11 February 2011

Citation: World Neurosurg. (2011) 76, 1/2:164-172.DOI: 10.1016/j.wneu.2011.02.029

Journal homepage: www.WORLDNEUROSURGERY.org

Available online: www.sciencedirect.com

1878-8750/$ - see front matter © 2011 Elsevier Inc.All rights reserved.

PEER-REVIEW REPORTS



This reprint is provided with the support of St. Jude Medical.

Documents

Clinical Safety of Brain Magnetic Resonance Imaging with ...mriquestions.com/uploads/3/4/5/7/34572113/dbs_review_complicati… · MRI parameters Use only 1.5-T horizontal-bore MRI