Reviews - Home | Diagnostic Radiology · 2018-04-05 · Johnson & Johnson, Medtronic, St Jude,...

Frank G. Shellock, PhDJohn V. Crues, MD

Index terms:Magnetic resonance (MR), biological

effects, ��.12142

Magnetic resonance (MR), safety,��.1214

Reviews

Published online before print10.1148/radiol.2323030830

Radiology 2004; 232:635–652

Abbreviations:DBS � deep brain stimulationFDA � Food and Drug

AdministrationRF � radiofrequencySAR � specific absorption rate

1 From the Keck School of Medicine,University of Southern California; andInstitute for Magnetic ResonanceSafety, Education, and Research, 7511McConnell Ave, Los Angeles, CA90045 (F.G.S.); and Radnet Manage-ment, Los Angeles, Calif (J.V.C.). Re-ceived May 21, 2003; revision re-quested July 18; revision receivedAugust 8; accepted October 8.Address correspondence to F.G.S.(e-mail: [email protected]).

F.G.S. maintains Web sites www.MRIsafety.com (with unrestricted edu-cational grant support by Bracco) andwww.IMRSER.org. The Institute forMagnetic Resonance Safety, Educa-tion, and Research and www.IMRSER.org are supported by unrestricted ed-ucational grants from Medtronic, Di-agnostic Village, Bracco, Esaote, Sie-mens Medical Solutions, GE MedicalSystems, Berlex Laboratories, Amer-sham Health, Magnetic ResonanceSafety Testing Services, Hitachi Medi-cal Systems America, Magmedix, Me-drad, Resonance Technology, InVivoResearch, Edwards Lifesciences, Bio-phan Technologies, Boston Scientific,MRI Devices, IGC-Medical Advances,Cook, Toshiba, Bard, and Philips Med-ical Systems. F.G.S. is a consultant toJohnson & Johnson, Medtronic, StJude, Guidant, Diagnostic Village,Bracco, Esaote, Siemens Medical Sys-tems, Hitachi Medical Systems Amer-ica, Magmedix, Edwards Lifesciences,Biophan Technologies, Boston Scien-tific, and Cook. J.V.C. is a consultant toMagneVu.2�� . Multiple body systems© RSNA, 2004

MR Procedures: BiologicEffects, Safety, and PatientCare1

The technology used for magnetic resonance (MR) procedures has evolved contin-uously during the past 20 years, yielding MR systems with stronger static magneticfields, faster and stronger gradient magnetic fields, and more powerful radiofre-quency transmission coils. Most reported cases of MR-related injuries and the fewfatalities that have occurred have apparently been the result of failure to followsafety guidelines or of use of inappropriate or outdated information related to thesafety aspects of biomedical implants and devices. To prevent accidents in the MRenvironment, therefore, it is necessary to revise information on biologic effects andsafety according to changes that have occurred in MR technology and with regardto current guidelines for biomedical implants and devices. This review provides anoverview of and update on MR biologic effects, discusses new or controversial MRsafety topics and issues, presents evidence-based guidelines to ensure safety forpatients and staff, and describes safety information for various implants and devicesthat have recently undergone evaluation.© RSNA, 2004

Magnetic resonance (MR) procedures have been utilized in the clinical setting for approxi-mately 20 years. During this time the technology has evolved continuously, yielding MRsystems with stronger static magnetic fields, faster and stronger gradient magnetic fields, andmore powerful radiofrequency (RF) transmission coils. Short-term exposures to the electro-magnetic fields used for MR procedures at the levels currently recommended by the U.S. Foodand Drug Administration (FDA) and with adherence to proper safety guidelines have yieldedrelatively few serious injuries for the more than 150 million MR examinations performed todate, with the exception of several second- and third-degree burns that have occurred (1–4).

Many of the MR-related injuries and the few fatalities that have occurred were theapparent result of failure to follow safety guidelines or of the use of inappropriate oroutdated information related to the safety aspects of biomedical implants and devices(1–7). The preservation of a safe MR environment requires constant attention to the careof patients and individuals with metallic implants and devices, because the variety andcomplexity of these objects constantly changes (5–7). Therefore, to guard against accidentsin the MR environment, it is necessary to revise information on biologic effects and safetyaccording to changes that have occurred in MR technology and with regard to the use ofcurrent guidelines for biomedical implants and devices (1,2,5–17).

In consideration of the above, this review will (a) provide an overview of and update on MRbiologic effects, (b) discuss new or controversial MR safety topics and issues, (c) presentevidence-based guidelines to ensure safety for patients and staff members, and (d) describe MRsafety information for various implants and devices that have recently undergone evaluation.

While a comprehensive discussion of MR biologic effects, safety, and patient care is notwithin the scope of this review, these topics have been addressed in recently publishedreview articles (8–12,16,17) and textbooks (5–7). In addition, there are at least two Websites devoted to MR safety that are updated with content on a frequent basis (18,19).

BIOLOGIC EFFECTS OF STATIC MAGNETIC FIELDS

The introduction of MR technology as a clinical imaging modality in the early 1980s isresponsible for a substantial increase in human exposure to strong static magnetic fields

Reviews

635

Ra

dio

logy

ESSENTIALS● Most reported cases of MR-related inju-

ries and the few fatalities that haveoccurred have apparently been the re-sult of failure to follow safety guidelinesor of use of inappropriate or outdatedinformation related to the safety as-pects of biomedical implants and de-vices.

● To prevent accidents and injuries in theMR environment, it is necessary to re-vise information on biologic effects andsafety according to changes that haveoccurred in MR technology and withregard to use of current guidelines forbiomedical implants and devices.

● The preservation of a safe MR environ-ment requires constant attention to thecare of patients and individuals withmetallic implants and devices, becausethe variety and complexity of these ob-jects constantly changes.

(1,9,16). Most MR systems in use todayoperate with magnetic fields rangingfrom 0.2 to 3.0 T. In the research setting,an exceptionally powerful MR system op-erating at 8.0 T is located at Ohio StateUniversity (Columbus). According to thelatest guidelines from the FDA, clinicalMR systems that use a static magneticfield up to 8.0 T are considered a “non-significant risk” for patients. The expo-sure of research subjects to fields strongerthan 8.0 T requires approval of the re-search protocol by an institutional re-view board and the informed consent ofthe subjects.

Schenck (1,9) conducted comprehen-sive reviews of biologic effects associatedwith exposure to static magnetic fields.With regard to short-term exposures (eg,limited exposures or those associated withthe clinical use of MR systems), the avail-able information for effects of static mag-netic fields on biologic tissues is exten-sive (1,9,20–38). Investigations includestudies on alterations in cell growth andmorphology, cell reproduction and ter-atogenicity, DNA structure and gene ex-pression, pre- and postnatal reproductionand development, blood-brain barrierpermeability, nerve activity, cognitivefunction and behavior, cardiovasculardynamics, hematologic indexes, temper-ature regulation, circadian rhythms, im-

mune responsiveness, and other biologicprocesses (20–38). In the majority ofthese studies, the authors concluded thatexposures to static magnetic fields pro-duce no substantial harmful biologic ef-fects. Although there have been some re-ports of potentially injurious effects ofstatic magnetic fields on isolated cells ororganisms, none of these effects havebeen verified or firmly established as ascientific fact (1,9). The relatively fewdocumented injuries that have occurredin association with MR system magnetswere attributed to the inadvertent pres-ence or introduction of ferromagnetic ob-jects (eg, oxygen tanks, aneurysm clips)into the MR environment (1,5–7,9).

With regard to the effects of long-termexposure to static magnetic fields, thereare interactions between tissues and staticmagnetic fields that could theoreticallylead to pathologic changes in human sub-jects (1,9,16). However, quantitative anal-ysis of these mechanisms indicates thatthey are below the threshold of impor-tance with respect to long-term adversebiologic effects (1,9,16).

At present, the pertinent literature doesnot contain carefully controlled studiesthat demonstrate the absolute safety ofchronic exposure to powerful magneticfields. With the increased clinical use ofinterventional MR procedures, there is acritical need for such investigations.However, it may be virtually impossibleto demonstrate “absolute safety,” giventhe various difficulties in conductingsuch a study. In addition, although thereis no evidence for a cumulative effect ofmagnetic field exposures on health, fur-ther studies of the exposed populations(eg, MR health care workers, patientswho undergo repeated studies) will behelpful in establishing rational guide-lines for occupational and patient expo-sures to static magnetic fields (1,9,16).

BIOLOGIC EFFECTS OFGRADIENT MAGNETIC FIELDS

During MR procedures, gradient mag-netic fields may stimulate nerves or mus-cles by inducing electric fields in pa-tients. This topic has been thoroughlyreviewed by Schaefer et al (8), Nyenhuiset al (39), and Bourland et al (40). Thepotential for interactions between gradi-ent magnetic fields and biologic tissue isdependent on a variety of factors, includ-ing the fundamental field frequency, themaximum flux density, the average fluxdensity, the presence of harmonic fre-quencies, the waveform characteristics of

the signal, the polarity of the signal, thedistribution of current in the body, theelectric properties, and the sensitivity ofthe particular cell membrane (8,39–48).

Several investigations have been con-ducted to characterize MR system–relatedgradient magnetic field–induced stimula-tion in human subjects (41–48). At suffi-cient exposure levels, peripheral nervestimulation is perceptible as a “tingling”or “tapping” sensation. At gradient mag-netic field exposure levels of 50%–100%above perception thresholds, patients maybecome uncomfortable or experience pain(8). At extremely high levels, cardiac stim-ulation is of concern. However, the in-duction of cardiac stimulation requiresexceedingly strong and/or rapid gradientmagnetic fields—more than an order ofmagnitude greater than those used in com-mercially available MR systems (8,39,40).Fortunately, current safety standards forgradient magnetic fields associated withpresent-day MR systems appear to pro-vide adequate protection from potentialhazards or injuries in patients (2,8,16,39).

Of interest, results of studies per-formed in human subjects indicate thatanatomic sites of peripheral nerve stimu-lation vary depending on the activationof a specific gradient (ie, x, y, or z gradi-ent) (8). Stimulation sites for x gradientsincluded the bridge of the nose, the leftside of the thorax, the iliac crest, the leftthigh, the buttocks, and the lower back.Stimulation sites for y gradients includedthe scapula, the upper arms, the shoul-der, the right side of the thorax, the iliaccrest, the hip, the hands, and the upperback. Stimulation sites for z gradients in-cluded the scapula, the thorax, the xy-phoid, the abdomen, the iliac crest, andthe upper and lower back (8). Typically,peripheral nerve stimulation sites were atbony prominences. According to Schaeferet al (8), because bone is less conductivethan the surrounding tissue it may in-crease current densities in narrow regionsof tissue between bone and skin, result-ing in lower nerve stimulation thresholdsthan expected.

ACOUSTIC NOISE

Various forms of acoustic noise are pro-duced in association with the operationof an MR system (49). The primary sourceof acoustic noise, however, is the gradi-ent magnetic field activated during theMR procedure. This noise occurs duringrapid alterations of current within thegradient coils that, in the presence of thepowerful static magnetic field of the MR

636 � Radiology � September 2004 Shellock and Crues

Ra

dio

logy

system, produce substantial (lorentzian)forces. Acoustic noise, manifested as loudtapping, knocking, or chirping sounds, isgenerated when these forces cause mo-tion or vibration of the gradient coils asthey impact their mountings.

Problems associated with acoustic noisefor patients and health care workers in-clude simple annoyance, difficulties inverbal communication, heightened anx-iety, temporary hearing loss, and, poten-tially, permanent hearing impairment(49–61). Acoustic noise may pose a par-ticular hazard to specific patient groupswho are at increased risk. Patients withpsychiatric disorders, the elderly, and pedi-atric patients may be confused or experi-ence heightened anxiety (49,51). Sedatedpatients may experience discomfort dueto high noise levels. Certain drugs areknown to increase hearing sensitivity(52). Neonates with immature anatomicdevelopment may have an increased re-action to acoustic noise, as has been re-ported by Philbin et al (53).

Characteristics of MR-relatedAcoustic Noise

Variations in MR-related acoustic noiseoccur with alterations in the gradient out-put (rise time or amplitude) associatedwith different MR parameters (49,54–64).Noise levels, pitch, and frequency char-acteristics are predominantly increasedwhen section thickness, field of view, rep-etition time, and echo time are decreased.The physical features of the MR system,especially the presence or absence of spe-cial sound insulation, and the materialand construction of gradient coils andsupport structures also affect the trans-mission of acoustic noise and its percep-tion by the patient.

The patient’s presence and the pa-tient’s size also affect the level of acousticnoise. An increase in acoustic noise hasbeen reported with a patient or volunteerpresent in the bore of the MR system(63); this may be due to pressure dou-bling (ie, an increase in sound pressure)close to an object, as sound waves reflectand undergo in-phase enhancement.Noise characteristics also have a spatialdependence. For example, noise levelshave been found to vary by as much as 10dB as a function of patient position alongthe magnet bore (63).

MR-related acoustic noise levels havebeen measured during a variety of pulsesequences for MR systems with staticmagnetic field strengths ranging from 0.2to 4.7 T (54–56,61–64). Recent studies per-formed with MR parameters that included

“worst-case” pulse sequences showed that,not surprisingly, fast gradient-echo, fast spin-echo, and echo-planar pulse sequences pro-duced the greatest acoustic noise levels(49,55,56).

MR-related Acoustic Noise andPermissible Limits

The FDA indicates that MR-related acous-tic noise levels must be below the level ofconcern established by pertinent federalregulatory or other recognized standards-setting organizations (2). If the acousticnoise is not below this level, the sponsor(ie, the manufacturer of the MR system)must recommend steps to reduce or alle-viate the noise perceived by the patient.A single upper limit of 140 dB is appliedto peak acoustic noise (2). However, theinstructions for use of MR systems mustadvise the MR system operator to providehearing protection to patients for opera-tion above an acoustic noise level of 99dB (2).

In general, acoustic noise levels re-corded by various researchers in associa-tion with conventional or routine MRprocedures have been below the maxi-mum limit permissible by the U.S. Occu-pational Safety and Health Administra-tion (2). Notably, when one considersthat the duration of exposure is one ofthe more important physical factors thatdetermine the effect of noise on hearing,then acoustic noise levels associated withMR procedures do not tend to be prob-lematic because of the relative short pe-riods of exposure (65,66).

Prevention of Acoustic NoiseProblems

Various techniques have been de-scribed to attenuate noise and, thus, pre-vent problems or hazards associated withexposure to MR-related acoustic noise(49,64). The simplest and least expensivemeans is to use disposable earplugs orcommercially available noise-abatementheadphones (49). Earplugs, when prop-erly used, can decrease noise by 10–30dB, which usually affords adequate pro-tection for MR environments with rela-tively loud MR systems. Regardless of thetechnique used, facilities operating withMR systems that generate substantialacoustic noise should require all patientsundergoing an examination to wear aprotective hearing device. Exposure ofstaff members, health care workers, andother individuals (eg, relatives, visitors) toloud MR systems is also of concern (49,56).Therefore, these individuals should like-

wise be required to use an appropriatemeans of hearing protection if they re-main in the room during the operation ofthese units (49).

BIOLOGIC EFFECTS OF RFFIELDS

The majority of the RF power transmittedfor MR imaging or spectroscopy (eg, car-bon decoupling, fast spin-echo pulse se-quences, magnetization transfer contrastpulse sequences) is transformed into heatwithin the patient’s tissues as a result ofresistive losses (11,67). Not surprisingly,the primary biologic effects associatedwith exposure to RF radiation are relatedto the thermogenic qualities of this elec-tromagnetic field (11,67–77).

Prior to 1985, there were no publishedreports concerning thermal or otherphysiologic responses of human subjectsexposed to RF radiation during MR pro-cedures. Since then, many investigationshave been conducted to characterize thethermal effects of MR procedure–relatedheating (68–74,78). This topic has beenreviewed by Schaefer (67,76) and Shel-lock (11).

MR Procedures and SpecificAbsorption Rate of RF Radiation

Thermoregulatory and other physio-logic changes that a human subject ex-hibits in response to exposure to RF radi-ation are dependent on the amount ofenergy that is absorbed. The dosimetricterm used to describe the absorption ofRF radiation is the specific absorptionrate (SAR) (11,67,76,79). The SAR is themass normalized rate at which RF poweris coupled to biologic tissue and is typi-cally expressed in watts per kilogram. Therelative amount of RF radiation that anindividual encounters during an MR pro-cedure is usually characterized with re-spect to the whole-body averaged andpeak SAR levels (ie, the SAR averaged in1 g of tissue).

Measurements or estimates of SAR arenot trivial, particularly in human sub-jects. There are several methods of deter-mining this parameter for the purpose ofRF energy dosimetry in association withMR procedures (67,76,79,80). The SARthat is produced during an MR procedureis a complex function of numerous vari-ables, including the frequency (ie, deter-mined by the strength of the static mag-netic field of the MR system), therepetition time, the type of RF coil used,the volume of tissue contained withinthe coil, the configuration of the ana-

Volume 232 � Number 3 Biologic Effects and Safety of MR � 637

Ra

dio

logy

tomic region exposed, and the orienta-tion of the body to the field vectors, aswell as other factors (11,67,76,79,80).

Thermophysiologic Responses toMR Procedure–related Heating

Thermophysiologic responses to MRprocedure–related heating depend onmultiple physiologic, physical, and envi-ronmental factors (11,67,76,77). Theseinclude the duration of exposure, the rateat which energy is deposited, the statusof the patient’s thermoregulatory system,the presence of an underlying healthcondition, and the ambient conditionswithin the MR system.

With regard to the thermoregulatorysystem, when subjected to a thermalchallenge the human body loses heat bymeans of convection, conduction, radia-tion, and evaporation. Each of thesemechanisms is responsible to a varyingdegree for heat dissipation as the bodyattempts to maintain thermal homeosta-sis (11,67,77,79). If the thermoregulatoryeffectors are not capable of totally dissi-pating the heat load, then there is anaccumulation, or storage, of heat alongwith an elevation in local and/or overalltissue temperatures (11,76,77).

Various underlying health conditionsmay affect an individual’s ability to tol-erate a thermal challenge, including car-diovascular disease, hypertension, diabe-tes, fever, old age, and obesity (81–85). Inaddition, medications such as diuretics,�-blockers, calcium blockers, amphetamines,muscle relaxants, and sedatives can alsogreatly alter thermoregulatory responsesto a heat load. In fact, certain medica-tions have a synergistic effect with re-spect to tissue heating if the heating isspecifically caused by exposure to RF ra-diation (86).

The environmental conditions that ex-ist in and around the MR system will alsoaffect the tissue temperature changes as-sociated with RF-induced heating. Dur-ing an MR procedure, the amount of tis-sue heating that occurs and the concomitantexposure to RF energy that is tolerable aredependent on environmental factors thatinclude ambient temperature, relativehumidity, and airflow.

MR Procedure–related Heating andHuman Subjects

To our knowledge, the first study ofhuman thermal response to RF radiation–induced heating during an MR procedurewas conducted by Schaefer et al (87). Tem-perature changes and other physiologic pa-

rameters were assessed in volunteer sub-jects exposed to relatively high, whole-body, averaged SARs (approximately 4.0W/kg). The data indicated that therewere no excessive temperature elevationsor other deleterious physiologic conse-quences related to these exposures to RFradiation (87).

Several studies were subsequently con-ducted with volunteer subjects and pa-tients undergoing clinical MR procedureswith the intent of obtaining informationthat would be applicable to patient pop-ulations typically encountered in the MRsetting (68–75). These investigations dem-onstrated that changes in body tempera-ture were relatively minor (ie, �0.6°C).While there was a tendency for statisti-cally significant increases in skin temper-atures to occur, these were of no seriousphysiologic consequence.

Of interest, various studies reported apoor correlation between body or skintemperature changes versus whole-bodyaveraged SARs during clinical MR proce-dures (69,73). These findings are not sur-prising considering the range of thermo-physiologic responses possible to a givenSAR that are dependent on the individu-al’s thermoregulatory system and thepresence of one or more underlying con-dition(s) that can alter or impair the abil-ity to dissipate heat.

An extensive investigation was con-ducted in volunteer subjects exposed to a1.5-T 64-MHz MR procedure with awhole-body averaged SAR of 6.0 W/kg(75), which, to our knowledge, is the high-est level of RF energy to which humansubjects have ever been exposed with anMR system. This excessive amount of RFradiation was achieved by using non-clinical MR imaging parameters (75).Tympanic membrane temperature, sixdifferent skin temperatures, heart rate,blood pressure, oxygen saturation, andskin blood flow were monitored (75). Thefindings indicated that an MR procedureperformed at a whole-body averaged SARof 6.0 W/kg can be physiologically toler-ated by an individual with normal ther-moregulatory function (75).

MR Procedure–related Heating andVery High Field Strength MRSystems

There are over 200 MR systems operat-ing with a static magnetic field strengthof 3 T, several operating at 4 T, a fewoperating at 7 T, one operating at 8 T(74), and at least one MR unit that oper-ates at a field strength higher than 8 T isin the final stage of installation (likely

completed by the time this article is pub-lished). For a given application, thesevery high field strength systems are capa-ble of generating RF power depositionsthat greatly exceed those associated witha 1.5-T MR system. Of note, with thedoubling of field strength (eg, 1.5 vs 3.0T), the RF power deposition increases fourtimes for a given MR imaging pulse se-quence. Therefore, investigations areneeded for characterization of thermalresponses in human subjects to deter-mine potential thermogenic hazards as-sociated with the use of these powerfulMR devices. To date, however, with theexception of work conducted at 8 T byKangarlu et al (74), there has been virtu-ally no investigation of MR procedure–related heating with regard to very highfield strength MR systems.

MR SAFETY AND PATIENTCARE

Screening Patients for MRProcedures and Individualsfor the MR Environment

The establishment of thorough and ef-fective screening procedures for patientsand other individuals is one of the mostcritical components of a program toguard the safety of all those preparing toundergo MR procedures or to enter theMR environment (5,13,15–17,89). An im-portant aspect of protecting individualsfrom MR system–related accidents andinjuries involves an understanding of therisks associated with the various im-plants, devices, accessories, and other ob-jects that may cause problems in this set-ting (5,13,15–17). This requires obtaininginformation and documentation aboutthese objects in order to provide the saf-est MR setting possible. In addition, be-cause MR-related incidents have beendue to deficiencies in screening methodsand/or a lack of proper control of accessto the MR environment (especially withregard to preventing personal items andother potentially problematic objectsfrom entering the MR room) (3,4), it iscrucial to set up procedures and guide-lines to prevent such incidents from oc-curring. Various guidelines and recommen-dations have been developed to facilitate thescreening process (15,17,88,89).

Screening patients for MR.—Certain as-pects of screening patients for MR proce-dures may take place during the schedul-ing process. This must be conducted by ahealth care worker who is speciallytrained in MR safety (17,88,89). That is,this individual should be (a) trained to


Ra

dio

logy

understand the potential hazards and is-sues associated with the MR environmentand MR procedures and (b) familiar withthe information contained on screeningforms for patients and individuals. Dur-ing this time, it may be ascertained if thepatient has any implant that may be con-traindicated for the MR procedure (eg, fer-romagnetic aneurysm clip, pacemaker) orif there is any condition that requires care-ful consideration (eg, patient is pregnantor has a disability). Preliminary screeninghelps to prevent scheduling of patientswho may be inappropriate candidates forMR examinations.

At the facility, it is advisable for everypatient to undergo comprehensive screen-ing in preparation for the MR examina-tion. Comprehensive patient screeninginvolves the use of a printed form to doc-ument the screening procedure, a reviewof the information on the screeningform, and an oral interview to verify theinformation and allow discussion of anyquestion or concern that the patient mayhave (15,88,89). A health care workertrained in MR safety must conduct thisaspect of patient screening. Variousforms have been developed for screeningpatients in preparation for MR proce-dures (5,15,17–19,88,89). An example ofa recently developed form for this use isshown in Figure 1 (18,19).

With the use of any type of writtenquestionnaire, limitations exist related toincomplete or incorrect answers pro-vided by the patient (18,19,88,89). Forexample, there may be difficulties associ-ated with patients who are impaired withrespect to their vision, language fluency,or level of literacy. Therefore, an appro-priate accompanying family member orother individual (eg, referring physician)should be involved in the screening pro-cess to verify any information that mayaffect patient safety. Versions of thisform should also be available in otherlanguages, as needed (ie, specific to thedemographics of the population servedby the MR facility) (17,88).

In the event that the patient is coma-tose or unable to communicate, the formshould be completed by the most quali-fied individual (eg, physician, familymember) with knowledge of the patient’smedical history and present condition. Ifthe screening information is inadequate,it is advisable to look for surgical scarson the patient and/or to obtain conven-tional radiographs of the skull and/orchest to search for implants that may beparticularly hazardous in the MR environ-ment (eg, aneurysm clip, cardiac pace-maker).

After completion of the screening formused for patients, a health care workertrained in MR safety must review thecontents of the form. Next, an oral inter-view should be conducted by the MRsafety–trained health care worker to ver-ify the information on the form and toallow discussion of any question or con-cern that the patient may have beforeundergoing the MR procedure. This al-lows for clarification or confirmation ofthe answers to the questions posed to thepatient so that there is no miscommuni-cation regarding important MR safety is-sues. In addition, because the patientmay not be fully aware of the medicalterminology used for a particular implantor device, it is imperative that this partic-ular information on the form be dis-cussed during the oral interview.

It should be noted that having under-gone a previous MR procedure withoutincident does not guarantee a safe subse-quent MR examination. Various factors(eg, static magnetic field strength of theMR system, orientation of the patient,orientation of a metallic implant or ob-ject) can substantially change the sce-nario (17,88,89). Therefore, a compre-hensive screening procedure must beconducted each time a patient preparesto undergo an MR procedure. This is notan inconsequential matter, because a sur-gical intervention or accident involving ametallic foreign body may have occurredthat could affect the safety of the patiententering the MR environment.

Screening individuals for the MR environ-ment.—Similar to the procedure con-ducted for screening patients, all otherindividuals (eg, MR technologists, pa-tient’s family members, visitors, alliedhealth professionals, maintenance work-ers, custodial workers, firefighters, secu-rity officers) should undergo screening byusing appropriate guidelines before beingallowed into the MR environment (17–19). This involves the use of a printedform to document the screening proce-dure, a review of the information on theform, and an oral interview to verify theinformation and allow discussion of anyquestion or concern that the individualmay have before entry to the MR envi-ronment is permitted.

In general, MR screening forms weredeveloped with patients in mind and,therefore, contain many questions thatare inappropriate or confusing to otherindividuals who may need to enter theMR environment. Therefore, a screeningform was recently created for individualswho need to enter the MR environmentand/or MR system room (Fig 2) (18,19).

To prevent problems that may occur inindividuals who respond to the MR fa-cility during emergencies, a procedureshould be in place to screen these indi-viduals well in advance of their entry tothe MR environment.

Metallic Orbital Foreign Bodiesand Screening

The single case report in 1986 by Kellyet al (90) about a patient who sustainedan ocular injury from a retained metallicforeign body has led to controversy re-garding the procedure required to screenindividuals prior to their entry to the MRenvironment (91–93). To date, this inci-dent is the only serious eye-related injurythat has occurred in association with theMR setting, according to recent review ofthe peer-reviewed literature and review ofdata files from the Manufacturer andUser Facility Device Experience Database(MAUDE; available at www.fda.gov/cdrh/maude.html) and the Medical Device Re-port (available at www.fda.gov/CDRH/mdrfile.html), both from the FDA Centerfor Devices and Radiological Health.

In the past, any individual or patientsuspected of having an orbital foreignbody typically underwent screening withconventional radiography of the orbits todetermine whether a metallic object waspresent. Thus, screening radiographs ofthe orbits were obtained routinely notonly in individuals who had a history ofinjury from a foreign body but also inthose who simply had a history of expo-sure to metallic objects, such as welders,grinders, metal workers, sculptors, andothers. Obviously, conventional radio-graphs of the orbits may have been ob-tained unnecessarily in many individualsbecause of this policy.

Seidenwurm et al (93) presented researchand a new set of guidelines for radiographicscreening of individuals suspected ofhaving metallic foreign bodies. Their in-vestigation addressed the cost-effective-ness of the use of a clinical versus a radio-graphic technique to screen individuals fororbital foreign bodies before an MR proce-dure (93). The costs of screening weredetermined on the basis of publisheddata, disability rating guides, and resultsof a practice survey. A sensitivity analysiswas performed for each variable. For theiranalysis, the benefit of screening was pre-vention of immediate, permanent, nona-meliorable, or unilateral blindness. Sei-denwurm et al (93) implemented thefollowing policy: “If a patient reports in-jury from an ocular foreign body that wassubsequently removed by a doctor or


Ra

dio

logy

Figu

re1.

Exam

ple

ofan

MR

pro

ced

ure

scre

enin

gfo

rmfo

rp

atie

nts

.(R

epri

nte

d,

wit

hp

erm

issi

on,

from

the

Inst

itu

tefo

rM

agn

etic

Res

onan

ceSa

fety

,Ed

uca

tion

,an

dR

esea

rch

.)


Ra

dio

logy

that resulted in negative findings on anyexamination, we perform MR imaging. . .Those persons with a history of injuryand no subsequent negative eye exami-nation are screened radiographically.”The findings of their study indicated thatthe use of clinical screening before radi-ography increased the cost-effectivenessof foreign body screening by an order ofmagnitude (ie, assuming base-case ocularforeign body removal rates). Of note isthat Seidenwurm et al have performedapproximately 100 000 MR proceduresusing this protocol without incident.

Thus, an occupational history of expo-sure to metallic fragments, by itself, isnot sufficient to mandate radiographicorbital screening (92,93). Therefore, cur-rent practice guidelines for foreign bodyscreening should be altered in consider-ation of this information and because ra-diographic screening before MR procedureson the basis of occupational exposure aloneis not clinically necessary, nor is it cost-effective (92,93).

Updated guidelines for orbital foreign bodyscreening.—The procedure to follow withregard to a patient suspected of having an

orbital foreign body involves a clinicalscreening protocol that entails asking thepatient if he or she has had an ocularinjury (93). If an ocular injury from ametallic object was sustained, the patientis asked if a medical examination wasconducted at the time of the injury and ifhe or she was informed by the doctorthat the object was completely removed(93). If (a) there was no injury, (b) theindividual was informed that the oph-thalmologic examination results werenormal, or (c) the foreign body was re-moved at the time of the injury, the pa-tient then proceeds to MR imaging. Onthe basis of the results of the clinicalscreening protocol, the patient should bescreened with conventional radiographyif an ocular injury related to a metallicobject was sustained and the patient wasnot informed that the postinjury eye ex-amination result was normal (93). In thiscase, the MR examination is postponedand the patient is scheduled for screen-ing radiography.

Excessive Heating and BurnsAssociated with MR Procedures

The use of RF coils, physiologic moni-tors, electronically activated devices, andexternal accessories or objects made fromconductive materials has caused exces-sive heating that resulted in burn injuriesto patients undergoing MR procedures(3–6,94–101). Heating of implants andsimilar devices may also occur in associ-ation with MR procedures, but this tendsto be problematic primarily for objectsmade from conductive materials that havean elongated shape, such as electrodes, leads,guidewires, and certain types of catheters(eg, catheters with thermistors or otherconducting components) (102–108).

More than 30 incidents of excessiveheating have been reported in patientsundergoing MR procedures in the UnitedStates that were unrelated to equipmentproblems or the presence of conductiveexternal or internal implants or materials(3,4,109). These incidents include first-,second-, and third-degree burns experi-enced by patients. In many of these cases,the reports pertaining to these incidentsindicated that the limbs or other bodyparts of the patients were in direct con-tact with body RF coils or other RF trans-mit coils of the MR systems or that therewere skin-to-skin contact points sus-pected to be responsible for these injuries(3,4,109).

In consideration of these injuries, guide-lines have been developed to prevent ex-cessive heating and burns related to MR

Figure 2. Example of a screening form for individuals who must enter an MR environment.(Reprinted, with permission, from the Institute for Magnetic Resonance Safety, Education, andResearch.)


Ra

dio

logy

procedures (Appendix A) (19). The adop-tion of these guidelines will help to en-sure that patient safety is maintained, es-pecially as more conductive materialsand electronically activated devices areused in association with MR procedures.

Tattoos and Permanent Cosmetics

Traditional (ie, decorative) and cosmetictattoo procedures have been performed forthousands of years. Cosmetic tattooing or“permanent cosmetics” are used to reshape,recolor, recreate, or modify eye shadow,eyeliner, eyebrows, lips, beauty marks,and cheek blush. In addition, permanentcosmetics are used to hide scars and forother aesthetic applications (110,111).

There is considerable controversy re-garding the MR safety aspects of tattoosand permanent cosmetics (112–121).Problems related to MR procedures andtattoos and permanent cosmetics are as-sociated with the use of iron oxide orother metal-based pigments. Because asmall number of patients with permanentcosmetics who underwent MR procedures(fewer than 10 documented cases) experi-enced transient skin irritation, cutaneousswelling, or heating sensations (3,4),many radiologists have refused to per-form MR procedures in individuals withpermanent cosmetics (Shellock FG, un-published observations, 2002). Obvi-ously, this undue concern for possibleadverse events prevents patients withpermanent cosmetics from having accessto a potentially important diagnostic im-aging modality (115).

In a study conducted by Tope andShellock (115), the frequency and sever-ity of adverse events associated with MRimaging were determined in a popula-tion of subjects with permanent cosmet-ics. A questionnaire was distributed toclients of cosmetic tattoo technicians.One hundred thirty-five (13.1%) studysubjects underwent MR imaging afterhaving permanent cosmetics applied. Ofthese, only two (1.5%) experienced prob-lems associated with MR imaging: Onesubject reported a sensation of “slighttingling” and the other subject reported asensation of “burning,” both transient innature (115). On the basis of these find-ings, as well as of other available infor-mation (3,4), it is apparent that MR pro-cedures may be performed in patientswith permanent cosmetics without anyserious soft-tissue reactions or adverseevents. Therefore, the presence of perma-nent cosmetics should not prevent pa-tients from undergoing MR procedures.

Of interest, decorative tattoos tend to

cause worse problems (including first-and second-degree burns) in patients un-dergoing MR procedures than do cos-metic tattoos. For example, Kreidstein etal (119) reported that a patient experi-enced a sudden burning pain at the siteof a decorative tattoo during MR imagingof the lumbar spine at 1.5 T. Surprisingly,in order to permit completion of the MRexamination, an excision of the tattooedskin was performed (119). The authors ofthis report stated, “Theoretically, the ap-plication of a pressure dressing of the tat-too may prevent any tissue distortiondue to ferromagnetic pull” (119). How-ever, this simple and relatively benignprocedure was not attempted in this pa-tient. The authors also indicated that “insome cases, removal of the tattoo may bethe most practical means of allowingMRI” (119). Kanal and Shellock (120)commented on this report in a letter tothe editor, suggesting that the responseto this situation was “rather aggressive.”Clearly, the trauma, expense, and mor-bidity associated with excision of a tattoofar exceed those that may be associatedwith MR-related tattoo interactions.

Because of the relatively remote possi-bility of an incident occurring in a pa-tient with permanent cosmetics or a tat-too and due to the relatively minor short-term complication or adverse event thatmay develop (ie, transient cutaneous red-ness and swelling) (3,4,115), the patientshould be permitted to undergo an MRprocedure. Any problem regarding per-formance of an MR procedure in a pa-tient with permanent cosmetics or a tat-too should not prevent the examination,because the diagnostic information thatis provided by this modality may be cru-cial for the care of the patient. For pa-tients in whom MR-related heating mayoccur, it is advisable to apply an ice packor cold compress to the site of the tattooor permanent cosmetics as a precaution-ary measure, since this a relatively innoc-uous procedure that adds little risk, timedelay, or expense to the MR examinationand could reduce the possibility of ther-mal injury (although, to date, there areno empiric data to support this).

Information on this topic has also beenprovided to patients by the FDA Center forFood Safety and Applied Nutrition, Officeof Cosmetics and Colors fact sheet (116),as follows: “The risks of avoiding an MRIwhen your doctor has recommended oneare likely to be much greater than therisks of complications from an interac-tion between the MRI and tattoo or per-manent makeup. Instead of avoiding anMRI, individuals who have tattoos or per-

manent makeup should inform the radi-ologist or technician of this fact in orderto take appropriate precautions, avoidcomplications, and assure the best re-sults.”

Pregnant Patients and MRProcedures

MR procedures have been used to eval-uate obstetric, placental, and fetal abnor-malities in pregnant patients for morethan 18 years (122–125). Initially, therewere substantial technical problems withthe use of MR imaging, due primarily tothe presence of image degradation causedby fetal motion. However, several tech-nologic improvements, including the de-velopment of high-performance gradientsystems and rapid pulse sequences, pro-vided advances that were especially use-ful for imaging pregnant patients. Thus,high-quality MR studies for obstetric andfetal applications may now be accom-plished routinely in the clinical setting (125).

Diagnostic imaging is often requiredduring pregnancy (122). Thus, it is notuncommon to consider the use of an MRprocedure in a pregnant patient. Safetyissues exist that are related to possibleadverse biologic effects associated withexposure to the static magnetic, gradientmagnetic, and RF electromagnetic fieldsused for MR procedures (5,13,122). Assuch, many laboratory and clinical re-search investigations have been con-ducted to determine the effects of the useof unenhanced MR procedures duringpregnancy (29,34–36,126,127). The over-all findings from these studies indicatethat there is no substantial evidence ofinjury or harm to the fetus; however, ad-ditional research on this topic is war-ranted.

Guidelines for MR in pregnant patients.—In1991, the Safety Committee of the Societyfor Magnetic Resonance Imaging issued adocument entitled “Policies, Guidelines,and Recommendations for MR ImagingSafety and Patient Management” (13),which stated that “MR imaging may beused in pregnant women if other non-ionizing forms of diagnostic imaging areinadequate or if the examination pro-vides important information that wouldotherwise require exposure to ionizingradiation (eg, fluoroscopy, computed to-mography). Pregnant patients should beinformed that, to date, there has been noindication that the use of clinical MR im-aging during pregnancy has produced del-eterious effects.” These guidelines havebeen subsequently adopted by the Amer-ican College of Radiology and are consid-


Ra

dio

logy

ered to be the standard of care with re-spect to the use of MR procedures inpregnant patients.

Accordingly, in cases where the refer-ring physician and attending radiologistcan defend that the findings of the MRprocedure have the potential to affect thecare of the mother or fetus (eg, to addressimportant clinical problems or help iden-tify potential complications, anomalies,or complex fetal disorders), the MR pro-cedure may be performed with oral andwritten informed consent, regardless ofthe trimester (13,122).

MR PROCEDURES ANDIMPLANTS AND DEVICES

The MR environment may be unsafe forindividuals with certain biomedical im-plants or devices, owing primarily tomovement or dislodgment of objectsmade from ferromagnetic materials (3–7,103,128–143). As previously stated,while excessive heating and the induc-tion of electric currents may also presentrisks to patients with implants or devices,these problems are typically associatedwith implants that have elongated con-figurations and/or are electronically acti-vated (eg, neurostimulation systems, car-diac pacemakers) (94,101–103).

To date, more than 1200 objects havebeen tested for MR safety, with over 200evaluated at 3 T or higher (5–7,18,128–142). This information is available to MRhealth care professionals and others aspublished reports, compiled lists, and, inits entirety, online at www.MRIsafety.com.The topic of MR safety for implants anddevices was recently reviewed by Shel-lock (5,103). As such, the intent for thematerial presented in the current reviewis to provide information for implants

and devices for which there may be con-troversy or confusion, with an update onobjects tested at 3 T or higher.

Evaluation of Implants and Devicesfor Safety in the MR Environment

The evaluation of an implant or devicewith regard to the MR environment isnot a trivial matter. The proper assess-ment of an object typically entails charac-terization of magnetic field interactions(translational attraction and torque), MR-related heating, induced electric currents,and artifacts. A thorough evaluation ofthe effects of the MR environment on thefunctional and operational aspects of cer-tain implants and devices may also benecessary. It is important to note that anobject demonstrated to be safe accordingto one set of MR conditions may be un-safe under more “extreme” conditions(eg, stronger static magnetic field, greaterlevel of RF power deposition, faster gra-dient field, different RF transmissioncoil). Accordingly, the specific test condi-tions for a given implant or device mustbe known before one makes a decisionregarding whether a particular object issafe for an individual in the MR environ-ment.

Magnetic field–related issues.—Magneticfield–related translational attraction andtorque are known to present hazards toindividuals with certain implants or de-vices (5–7). Currently, MR systems usedin clinical and research settings operatewith a static magnetic field that rangesfrom 0.2 to 8.0 T. Most previous ex vivotests performed to assess objects for MRsafety used units with a static magneticfield of 1.5 T or lower (5,103). Accord-ingly, this could present problems, inso-far as it is possible that an object that

displayed “weakly” ferromagnetic quali-ties in association with a 1.5-T MR systemmay exhibit substantial magnetic fieldinteractions with an MR system operat-ing at a stronger static magnetic fieldstrength (5,103,128–131). Therefore, in-vestigations have been conducted andare ongoing in which 3- and 8-T MR sys-tems are being used to determine MRsafety regarding implants and devices rel-ative to these powerful units (128–131).This is especially crucial because most fa-cilities with a 3-T MR imager currently donot perform MR procedures in patientswith metallic objects because of the lackof safety information.

Long-bore versus short-bore MR systems.—Different magnet configurations exist forcommercially available 1.5- and 3.0-T MRsystems. These include conventional“long-bore” and “short-bore” systemsused for whole-body (1.5- and 3.0-T MRsystems) and head-only (3.0-T MR sys-tems) clinical applications. In recent re-ports, it has been indicated that short-boreMR systems have significantly higher spa-tial gradients than do long-bore MR sys-tems, especially for MR systems operatingat 3 T (129,130). This can affect MR safetyfor a given metallic implant or device(129,130). Therefore, this is an additionalfactor that must be taken into consider-ation when evaluating objects for safetyin the MR environment.

Aneurysm clips.—The presence of an in-tracranial aneurysm clip (Fig 3) in a pa-tient referred for an MR procedure or inan individual who needs to enter the MRenvironment represents a situation thatrequires careful consideration because ofthe associated risks (5–7,103,137–152).Aneurysm clips made from ferromagneticmaterials are contraindicated for MR pro-cedures because excessive magneticallyinduced forces may displace these clips,causing serious injury or death. By com-parison, aneurysm clips classified as non-ferromagnetic or weakly ferromagnetic(eg, made from Elgiloy, Phynox, titaniumalloy, or commercially pure titanium)have been tested and shown to be safe forpatients undergoing MR procedures at1.5 T or lower (5–7,137–152). In 1998,Shellock and Kanal (146) provided guide-lines based on the relevant peer-reviewedliterature for the care of a patient with ananeurysm clip (Appendix B).

Various studies have been performedto support imaging in patients with non-ferromagnetic aneurysm clips (Fig 4). Forexample, Pride et al (145) reported find-ings from several patients with nonferro-magnetic aneurysm clips who were im-aged at 1.5 T. There was no objective

Figure 3. Examples of aneurysm clips with a variety of shapes andsizes (different versions of Spetzler Titanium Aneurysm Clips; Health-care Corporation, V. Mueller Neuro/Spine, San Carlos, Calif). Aneu-rysm clips may be made from various materials, including ferromag-netic and nonmagnetic metals.


Ra

dio

logy

adverse outcome for these patients, whichconfirmed that MR procedures can beperformed safely in patients with nonfer-romagnetic clips. Brothers et al (138) alsodemonstrated MR safety at 1.5 T for pa-tients with nonferromagnetic aneurysmclips. Their report was particularly impor-tant, because MR imaging was found tobe better than computed tomography forpostoperative assessment of patientswith aneurysms, especially with regard tothe ability to show small zones of isch-emia (138).

To our knowledge, only one ferromag-netic aneurysm clip–related fatality hasbeen reported in the peer-reviewed liter-ature (143). This incident was the resultof erroneous information pertaining tothe type of aneurysm clip that waspresent in the patient—the clip was be-lieved to be a nonferromagnetic Yasargilaneurysm clip (Aesculap, South San Fran-cisco, Calif) but turned out to be a ferro-magnetic Vari-Angle clip (Codman &Shurtleff, Raynham, Mass) (143).

Aneurysm clips tested at 3 and 8 T.—Various aneurysm clips have been testedfor magnetic field interactions in associ-ation with 3- and 8-T MR systems (128–130). Findings indicated that the clipseither exhibited a lack of magnetic fieldinteractions or relatively weak magneticfield–related translational attraction andtorque at 3 T. Accordingly, some aneu-rysm clips are considered to be entirelysafe for patients undergoing procedureswith MR systems operating at 3 T, whileothers require further characterization ofmagnetic field–induced torque (128,129).

An early investigation to determinemagnetic field interactions for medicalimplants at 8 T involved an assessment ofaneurysm clips (131). Aneurysm clipsrepresentative of those made from non-ferromagnetic or weakly ferromagneticmaterials used for temporary or perma-nent treatment of aneurysms or arterio-venous malformations were selected forthat study. Test results showed that MRsafety at 8 T for the aneurysm clips wasdependent not only on the material butalso on the dimensions, model, shape,size, and blade length of a given clip.

Heart valve prostheses and annuloplastyrings.—Numerous heart valve prosthesesand annuloplasty rings have undergonetesting for MR safety (5–7,128,153–158).Of these, the majority showed measur-able but relatively minor translational at-traction and/or torque in associationwith exposure to the MR systems used fortesting. Since the magnetic field–relatedforces exerted on heart valves and an-nuloplasty rings are deemed minimal

compared with the force exerted by thebeating heart (ie, approximately 7.2 N)(153,154), an MR procedure is consideredto be safe for a patient with any of theheart valve prostheses or annuloplastyrings that have undergone testing to date(5–7,128,153–158). This includes theStarr-Edwards model Pre-6000 heartvalve prosthesis, which had previouslybeen suggested to be potentially hazard-ous for a patient in the MR environment.

Heart valve prostheses and annuloplastyrings tested at 3 T.—Many heart valveprostheses and annuloplasty rings havenow been evaluated for MR safety by us-ing 3-T units (128). Findings indicate thatone annuloplasty ring (Carpentier-Ed-wards Physio Annuloplasty Ring, Mitralmodel 4450; Edwards Lifesciences, Ir-vine, Calif) showed relatively minor mag-netic field interactions. Therefore, similarto heart valve prostheses and annulo-plasty rings tested at 1.5 T, because theactual attractive forces exerted on theseimplants are deemed minimal comparedto the force exerted by the beating heart,MR procedures at 3 T are not consideredto be hazardous for individuals withthese implants (5,128).

Additional heart valves and annulo-plasty rings from the Medtronic HeartValve Division (Minneapolis, Minn) have

undergone MR safety testing at 3 T. Theseimplants were tested for magnetic fieldinteractions and artifacts by using ashielded 3-T MR system. According to in-formation provided by Medtronic (BayerKM, personal communication, 2002),these specific implants are safe for pa-tients undergoing procedures with MRsystems operating up to 3 T.

Coils, filters, and stents.—There are manydifferent types of coils, filters, and stentsthat are used for a variety of applications(Fig 5). These implants are commonlymade from metallic materials such as plat-inum, titanium, stainless steel, Phynox,Elgiloy, and nitinol, which are mostlynonmagnetic or weakly ferromagnetic at1.5 T or lower (5,159–169). Heating andinduced currents have been evaluated fora wide variety of shapes and sizes of theseimplants, and there do not appear to beany safety issues for these devices. Forthose coils, filters, and stents found tohave no magnetic field interactions, anMR procedure may be performed imme-diately after placement (5–7,159,160).However, for those implants made fromweakly ferromagnetic materials, it is typ-ically recommended to wait 6–8 weeks to

Figure 4. Transverse T1-weighted spin-echoMR image (repetition time, 500 msec; echotime, 20 msec) of the brain obtained at 1.5 T ina patient with nonferromagnetic aneurysmclips. Note the presence of relatively small sig-nal void artifacts (arrowheads) associated withthese implants.

Figure 5. Examples of (a) an intravascular fil-ter (Recovery Nitinol Filter; Bard PeripheralVascular, Tempe, Ariz) and (b) stents (Endo-med, Phoenix, Ariz) that have undergone MRsafety testing.


Ra

dio

logy

allow tissue ingrowth to help retain theimplant in place (5–7,159,160). If there isany possibility that a coil, filter, or stent isnot positioned properly or is not firmlyin place, the patient should not be al-lowed into the MR environment.

It should be noted that because coils,filters, and stents are being developed onan ongoing basis, general MR safetyguidelines cannot be provided for theseimplants. Therefore, it is necessary to ob-tain documentation that clearly identi-fies the device, material, and manufacturerin order to avoid hazardous situations inthe MR environment. The results of astudy by Taal et al (169) support the factthat not all stents are safe for patientsundergoing MR procedures. They re-ported that “an appreciable attractionforce and torque” was found for twotypes of Gianturco stents. In consider-ation of these results, Taal et al advisedthat “specific information on the type ofstent is necessary before a magnetic reso-nance imaging examination is planned.”

MR safety at 3 T and coils and stents.—Several different coils and stents havebeen evaluated at 3 T (103,161). Of theimplants tested, two displayed magneticfield interactions that exceeded levelsthat might present risks to patients (103).However, similar to other coils andstents, tissue ingrowth may be sufficientto prevent these implants from posing asubstantial risk to a patient or individualin the 3-T MR environment. Thus, thisMR safety issue warrants further study.

Essure device.—The Essure device (Con-ceptus, San Carlos, Calif) is an implantdeveloped for permanent female contra-ception (170). It is composed of 316Lstainless steel, platinum, iridium, nickel-titanium alloy, silver solder, and polyeth-ylene terephthalate fibers. The Essuredevice is a dynamically expanding micro-coil that is placed in the proximal sectionof the fallopian tube by using a noninci-sional technique. Subsequently, the Es-sure device elicits a benign tissue re-sponse resulting in tissue in-growth thatanchors it and occludes the fallopiantube, resulting in permanent contracep-tion.

An MR safety assessment of this im-plant involved testing for magnetic fieldinteractions at 1.5 T, heating, inducedelectric currents, and artifacts (170). Thefindings indicated that it is safe for a pa-tient with the Essure device to undergoan MR procedure with an MR system op-erating at 1.5 T or lower.

Essure device and testing at 3 T.—TheEssure device was recently evaluated forMR safety at 3 T and was found to be safe

for patients undergoing MR proceduresoperating at this field strength (128).

TheraSeed radioactive seed implant.—The TheraSeed radioactive seed implant(Theragenics, Buford, Ga) is used to de-liver low-level radiation from palladium103 to the prostate gland to treat cancer.This relatively small implant is composedof a titanium tube with two graphite pel-lets and a lead marker inside. Treatmentmay involve placement of 80–120 seeds.MR testing for magnetic field interac-tions, heating, induced currents, and ar-tifacts revealed that the TheraSeed im-plant is safe for patients undergoing MRprocedures at 1.5 T or lower.

Cardiac pacemakers.—Cardiac pacemak-ers are the most common electronically ac-tivated implants found in patients re-ferred for MR procedures. Unfortunately,the presence of a pacemaker is consideredto be a strict contraindication for the MRenvironment (5–7,169–171). Potentialadverse interactions between pacemakersand MR procedures include movement ofthe pulse generator or leads, electrodeheating, induction of ventricular fibrilla-tion, rapid pacing, reed switch malfunc-tion (or normal reed switch function inthe presence of a powerful magnetic field),asynchronous pacing, inhibition of pacingoutput, alteration of programming withpossible damage to pacemaker circuitry,and other problems (5–7,171–190). Someof these issues are theoretic, while othershave been studied in vitro, in laboratoryanimals, and in human subjects.

More than 10 deaths have been attrib-uted to MR procedures performed in pa-tients with a cardiac pacemaker (3,4,189,190). These fatalities were poorly character-ized, since there was no electrocardio-graphic monitoring during the examina-tions. Furthermore, for each case, themode of death (ie, mechanism responsi-ble for the adverse cardiac pacemaker–MR procedure interaction) was not re-ported, and it was unknown whetherthese patients were pacemaker depen-dent (3,4,189,190). Of importance, therehave been no deaths associated with phy-sician-supervised imaging (189,190). In arecent letter to the editor addressing thecontroversy that exists with regard to im-aging patients with cardiac pacemakers,Gimbel (190) pointed out that pace-maker-related deaths occurred in patients“‘inadvertently’ placed in the MRI envi-ronment without the attending physi-cian conducting the MRI knowing thatthe patient being scanned had a pace-maker. Thus, none of the easily imple-mented techniques that might have al-

lowed a harmless scan to proceed wereimplemented.”

To date, more than 200 patients with acardiac pacemaker have undergone MRprocedures safely, either inadvertently orduring purposeful monitored attemptsto perform much-needed examinations(179,180,184,187–191). Thus, there isgrowing evidence that MR examinationsmay be performed in certain patients byfollowing highly specific procedures andMR conditions. Accordingly, restrictionsfor conducting MR procedures in pa-tients with cardiac pacemakers may bemodified in the near future. Until then, itis advisable to continue to restrict all pa-tients with cardiac pacemakers from theMR environment.

Investigations in human subjects withcardiac pacemakers have suggested variousstrategies for safe MR procedures. Thesestrategies include imaging only non–pace-maker-dependent patients, programmingthe pacemaker device to an “off” or asyn-chronous mode, programming to a bipo-lar lead configuration, limiting the RF en-ergy, and performing MR examinationsonly if the pulse generator is positionedoutside of the bore of the MR system(179,180,184,185,187,188).

In a recent study by Martin et al (191),however, results of MR performed at 1.5T indicate that these strategies may notbe necessary for non–pacemaker-depen-dent patients at 1.5-T MR imaging. Intheir investigation, in order to examinerisk in the broadest possible population,no restrictions were placed on the anat-omy imaged, the type of pulse sequenceand imaging parameters used for MR im-aging, or the type of pacemaker presentin the patient. Pacemaker-dependent pa-tients were excluded to eliminate problemsif pacing was inhibited during imaging. Ofimportance, absolute requirements forperforming MR procedures in these non–pacemaker-dependent patients includedthe attendance of a cardiologist withpacemaker expertise, the presence of re-suscitation equipment in proximity tothe MR system room, and the presence ofa physician certified in advanced cardiaclife support who could respond to anyuntoward consequence. As such, it is im-portant to recognize that imaging thesenon–pacemaker-dependent patients wasnot a trivial matter and required contin-uous monitoring and the means to rap-idly intervene in the event of an emer-gency (191).

Findings from the study by Martin et al(191) showed that 1.5-T MR proceduresdid not cause substantial problems or dif-ficulties. Furthermore, the results of this


Ra

dio

logy

investigation emphasized that it was notnecessary to inhibit the pacing pulse, toreprogram the pulse generator, or tochange MR parameters to achieve safety,as was done in prior studies in patientswith cardiac pacemakers. However, giventhe infinite possibilities of pacing sys-tems and cardiac and lead geometry, aswell as variable RF and gradient magneticfields, absolute safety with regard to pace-maker and MR interactions cannot be as-sured under all operational conditions.Nevertheless, on the basis of informationin the peer-reviewed literature it appearsthat with appropriate patient selection, aswell as continuous monitoring and pre-paredness for resuscitation efforts, perfor-mance of MR procedures in patients withan implanted cardiac pacemaker but whoare not pacemaker dependent may beachieved with reasonable safety, even atstatic magnetic field strengths of 1.5 T.

In the past, the presence of any elec-tronically activated implant was consid-ered a strict contraindication for an indi-vidual in the MR environment. Over theyears, however, various studies have beenperformed to define safety criteria forelectronic devices (104,106–108). There-fore, if highly specific guidelines are fol-lowed, MR procedures may be conductedsafely in patients with various electroni-cally activated implants, including neuro-stimulation systems, cochlear implants,and programmable drug infusion pumps(5–7,104,106–108). In fact, some of theseelectronically activated devices have re-ceived approval from the FDA for “MRsafe” labeling claims.

In consideration of the findings forconducting safe MR procedures in pa-tients with electronically activated de-vices that have been published in thepeer-reviewed literature, it is hoped thatcardiac pacemaker manufacturers willbe encouraged to proactively supportand/or conduct investigations directedtoward identifying safety criteria for theirrespective devices. This will ultimatelyhave a substantial effect on patient careand the overall health care of patientswith pacemakers who may require MRprocedures.

Neurostimulation system for deep brainstimulation.—Because of the increased in-terest in the use of deep brain stimulation(DBS) of the thalamus, globus pallidus,and subthalamic nucleus for treatment ofmedically refractory movement disordersand other types of neurologic conditions,the number of patients receiving im-plantable pulse generators and DBS elec-trodes is rapidly growing (106,107,192–194). The use of MR imaging in patients

with neurostimulation systems is fre-quently desired for surgical planning, aswell as for the ongoing management ofunderlying conditions (106,107,192–195).In addition, MR imaging may be neededin various clinical scenarios, includingverification of lead position, evaluationof patients with poor or worsening out-come, and examination of patients withother pathologic abnormalities unrelatedto DBS neurostimulation, such as stroke,tumor, or hemorrhage (107).

As with all electronically activated de-vices in the MR environment, it is gener-ally recommended that patients with aneurostimulation system should not un-dergo MR imaging because of the poten-tial for serious consequences, includingmovement of the leads or implantablepulse generator, excessive MR imaging–related heating, induced electric currents,and functional disruption of the opera-tional aspects of the device (5–7). Thus,before performing MR in a patient with aDBS system, it is essential to collect invitro experimental data to define MRconditions that may permit imaging tobe performed safely (106,107).

From an MR safety point of view, thegreatest concern for electronically acti-vated or electrically conductive implantsin the brain is excessive MR imaging–related heating, which can cause irrevers-ible tissue damage (106,107). Resultsfrom studies conducted to date (106,107)and a recent report (195) revealed thatthere is a realistic potential for injury dueto excessive MR imaging–related heatingof neurostimulation systems used forDBS.

Recently, investigators have evaluatedMR-related heating for the only neuro-stimulation system (Activa Tremor Con-trol System; Medtronic) approved by theFDA for use in chronic DBS (106,107).This neurostimulation system is a fullyimplantable multiprogrammable devicedesigned to deliver electric stimulation tothe thalamus or other brain structures.The basic implantable system is com-posed of the neurostimulator (or im-plantable pulse generator), the DBS lead,and an extension that connects the leadto the implantable pulse generator. Thisneurostimulation system delivers high-frequency electric stimulation to a multi-ple-contact electrode placed in the ven-tral intermediate nucleus of the thalamusor another anatomic site.

In their studies on neurostimulationsystems, Rezai et al (106) and Finelli et al(107) indicated that MR safety for neuro-stimulation systems is highly dependenton a number of critical factors. To simu-

late a worst-case clinical application ofDBS, these investigations evaluated bilat-eral DBS applications, such that two neuro-stimulators, two extensions, and two leadswere assessed during in vitro experiments.Different configurations were evaluated forthe bilateral neurostimulation systems tocharacterize worst-case and clinically rel-evant positioning scenarios (106,107).MR imaging procedures were performedon a gel-filled phantom designed to ap-proximate the head and upper torso of ahuman subject. Temperature changeswere studied in association with MR ex-aminations conducted at 1.5 T and 64MHz at various levels of RF energy byusing the transmit-receive RF body coiland transmit-receive RF head coil. Thefindings from these studies indicated thatsubstantial heating occurs under certainconditions, while other conditions pro-duced relatively minor physiologicallyinconsequential temperature increases.Furthermore, factors that strongly influ-enced local temperature increases at theelectrode tip included the positioning ofthe neurostimulation system (especiallythe electrode), the type of RF coil used,and the SAR used for the MR procedure.

According to the study by Rezai et al(106), MR-related heating does not ap-pear to present a major safety concern forpatients with the bilateral neurostimula-tion systems that underwent testing, aslong as highly specific guidelines pertain-ing to the positioning of these neuro-stimulation devices and to the parame-ters used for MR imaging are carefullyadhered to. Finelli et al (107) reportedthat MR imaging sequences commonlyused for clinical procedures can be per-formed safely with the use of a transmit-receive RF head coil at 1.5 T in patientswith a bilateral DBS system.

It should be noted that most present-day high-field-strength MR systems areused with a body coil to transmit RF anda receive-only head coil. Therefore, addi-tional studies are required to characterizethe effect of the use of this transmit-re-ceive RF coil combination with regard toMR imaging–related heating of neuro-stimulation systems used for DBS.

It is important to note that the exactsafety recommendations for the particu-lar neurostimulation system, with regardto the pulse generator, leads, electrodes,operational conditions for the device, po-sitioning of these components, and MRsystem conditions, must be carefully fol-lowed for MR imaging (106,107,195). Ashighlighted by two recent serious acci-dents (195), the failure to follow safetyrecommendations strictly may result in


Ra

dio

logy

serious temporary or permanent injury tothe patient, including the possibility oftransient dystonia, paralysis, coma, or evendeath.

Gastric electric stimulation.—Gastric elec-tric stimulation, performed by using a spe-cialized neurostimulation device (EnterraTherapy Gastric Electrical StimulationSystem; Medtronic), is indicated for treat-ment of patients with chronic intractablenausea and vomiting secondary to gas-troparesis of diabetic or idiopathic origin.Gastric electric stimulation uses mildelectric pulses to stimulate the stomachto help control symptoms associatedwith gastroparesis.

The gastric electric stimulation deviceis composed of a neurostimulator, an im-plantable intramuscular lead, and an ex-ternal programming system. Currently,the use of MR procedures in patients withthis device is contraindicated owing topossible hazards related to dislodgmentor heating of the neurostimulator and/orthe leads used for stimulation. In addi-tion, the voltage induced through thelead and neurostimulator may cause un-comfortable “jolting” or “shocking” lev-els of stimulation (5).

Postoperative MR Procedures

Because confusion exists regarding theissue of performing an MR procedureduring the postoperative period in a pa-tient with a metallic implant or device,guidelines have been developed pertain-ing to this MR safety topic (19). Studyresults have supported that, if a metallicobject is a passive implant (ie, there is noelectronically or magnetically activatedcomponent associated with the opera-tion of the object) and it is made from anonferromagnetic material (eg, titanium,titanium alloy, nitinol), the patient withthe object may undergo an MR procedureat 1.5 T or lower immediately after im-plantation (5–7,135,145,149,159). In fact,there are several reports that describeplacement of vascular stents and other im-plants with MR-guided procedures that in-clude the use of high-field-strength (1.5-T)MR systems (164,167,168). In addition, apatient or individual with a nonferromag-netic passive implant would be allowed toenter the MR environment associated witha 1.5-T or lower-strength MR system imme-diately after implantation of such an ob-ject. Currently, there are few data to pro-vide guidelines for MR environments withimagers operating at 3 T or higher.

For an implant or device that exhibitsweakly magnetic qualities, it is typicallynecessary to wait for 6–8 weeks after im-

plantation before performing an MR pro-cedure or allowing the individual to enterthe MR environment (5–7,156,160). Forexample, certain intravascular and intra-cavitary coils, stents, filters, and cardiacoccluders designated as weakly ferro-magnetic become firmly incorporatedinto tissue 6–8 weeks after placement. Inthese cases, retentive forces, or counter-forces, provided by tissue ingrowth, scar-ring, or granulation essentially serve toprevent these objects from presentinghazards to individuals in the MR envi-ronment. Those implants or devices thatmay be weakly magnetic but that are rig-idly fixed in the body, such as a bonescrew, may be studied immediately in thepostoperative period (19). Typically, spe-cific information pertaining to the rec-ommended postoperative waiting periodmay be found in the labeling or productinsert for a weakly magnetic implant ordevice.

If there is any concern regarding theintegrity of the tissue with respect to itsability to retain the implant or object inplace or if the implant cannot be prop-erly identified, the individual in suchcases should not be exposed to the MRenvironment. Specific information per-taining to the recommended postopera-tive waiting period may be found in thelabeling or product insert for a weaklymagnetic implant or device.

CONCLUSIONS

With the continued advances in MRtechnology and the development ofmore sophisticated implants and devices,there is an increased potential for hazard-ous situations to occur in the MR envi-ronment. Therefore, to prevent incidentsand accidents, it is necessary to be awareof the latest information pertaining toMR biologic effects, to use current evi-dence-based guidelines to ensure safetyfor patients and staff members, and tofollow proper recommendations pertain-ing to biomedical implants and devices.

APPENDIX A

The following guidelines, reprinted, withpermission, from the Institute for MagneticResonance Safety, Education, and Research(19), pertain to the prevention of excessiveheating and burns in association with MRprocedures.

1. Prepare the patient for the MR proce-dure by ensuring that there are no unnec-essary metallic objects contacting the pa-tient’s skin (eg, metallic drug delivery patches,jewelry, necklaces, bracelets, key chains, etc).

2. Prepare the patient for the MR proce-dure by using insulation material (ie, appro-priate padding) to prevent skin-to-skin con-tact points and the formation of “closed-loops” from touching body parts.

3. Insulating material (minimum recom-mended thickness, 1-cm) should be placedbetween the patient’s skin and transmit RFcoil that is used for the MR procedure (al-ternatively, the RF coil itself should be pad-ded). For example, position the patient sothat there is no direct contact between thepatient’s skin and the body RF coil of theMR system. This may be accomplished byhaving the patient place his/her arms overhis/her head or by using elbow pads or foampadding between the patient’s tissue andthe body RF coil of the MR system. This isespecially important for those MR examina-tions that use the body coil or other large RFcoils for transmission of RF energy.

4. Use only electrically conductive devices,equipment, accessories (eg, ECG leads, elec-trodes, etc), and materials that have beenthoroughly tested and determined to besafe and compatible for MR procedures.

5. Carefully follow specific MR safety cri-teria and recommendations for implantsmade from electrically conductive materials(eg, bone fusion stimulators, neurostimula-tion systems, etc).

6. Before using electrical equipment,check the integrity of the insulation and/orhousing of all components including sur-face RF coils, monitoring leads, cables, andwires. Preventive maintenance should bepracticed routinely for such equipment.

7. Remove all non-essential electricallyconductive materials from the MR system(ie, unused surface RF coils, ECG leads, ca-bles, wires, etc).

8. Keep electrically conductive materialsthat must remain in the MR system fromdirectly contacting the patient by placingthermal and/or electrical insulation be-tween the conductive material and the pa-tient.

9. Keep electrically conductive materialsthat must remain within the body RF coil orother transmit RF coil of the MR systemfrom forming conductive loops. Note: Thepatient’s tissue is conductive and, there-fore, may be involved in the formation of aconductive loop, which can be circular, U-shaped, or S-shaped.

10. Position electrically conductive mate-rials to prevent “cross points.” For example,a cross point is the point where a cablecrosses another cable, where a cable loopsacross itself, or where a cable touches eitherthe patient or sides of the transmit RF coilmore than once. Even the close proximityof conductive materials with each othershould be avoided because some cables andRF coils can capacitively couple (withoutany contact or crossover) when placed closetogether.


Ra

dio

logy

11. Position electrically conductive mate-rials to exit down the center of the MRsystem (ie, not along the side of the MRsystem or close to the body RF coil or othertransmit RF coil).

12. Do not position electrically conduc-tive materials across an external metallicprosthesis (eg, external fixation device, cer-vical fixation device, etc) or similar devicethat is in direct contact with the patient.

13. Allow only properly trained individu-als to operate devices (eg, monitoring equip-ment) in the MR environment.

14. Follow all manufacturer instructionsfor the proper operation and maintenanceof physiologic monitoring or other similarelectronic equipment intended for use dur-ing MR procedures.

15. Electrical devices that do not appearto be operating properly during the MR pro-cedure should be removed from the patientimmediately.

16. Closely monitor the patient duringthe MR procedure. If the patient reportssensations of heating or other unusual sen-sation, discontinue the MR procedure im-mediately and perform a thorough assess-ment of the situation.

17. RF surface coil decoupling failures cancause localized RF power deposition levelsto reach excessive levels. The MR systemoperator will recognize such a failure as a setof concentric semicircles in the tissue onthe associated MR image or as an unusualamount of image nonuniformity related tothe position of the RF coil.

APPENDIX B

The following guidelines, adapted from areport by Shellock and Kanal (146), pertainto the care of a patient with an aneurysmclip referred for an MR procedure.

1. Specific information (ie, manufacturer,type or model, material, and lot and serialnumbers) about the aneurysm clip must beknown so that only patients with nonferro-magnetic or weakly ferromagnetic clips areallowed into the MR environment. This in-formation is provided in the labeling of theaneurysm clip by the manufacturer. The im-planting surgeon is responsible for properlycommunicating this information in the pa-tient’s records.

2. An aneurysm clip that is in its originalpackage and is made from Phynox, Elgiloy,MP35N, titanium alloy, commercially puretitanium, or other material known to benonferromagnetic or weakly ferromagneticdoes not need to be evaluated for ferromag-netism; as such, these are considered to besafe for MR procedures performed at 1.5 Tor lower.

3. The radiologist and implanting surgeonare responsible for (a) evaluating the infor-mation pertaining to the aneurysm clip,

(b) verifying its accuracy, (c) obtaining writ-ten documentation, and (d) deciding to per-form the MR procedure after consideringthe risk-versus-benefit aspects for a givenpatient.

References1. Schenck JF. Health effects and safety of

static magnetic fields. In: Shellock FG,ed. Magnetic resonance procedures:health effects and safety. Boca Raton,Fla: CRC, 2001; 1–30.

2. Zaremba LA. FDA guidance for magneticresonance system safety and patient ex-posures: current status and future con-siderations. In: Shellock FG, ed. Magneticresonance procedures: health effects andsafety. Boca Raton, Fla: CRC, 2001; 183–196.

3. U.S. Food and Drug Administration,Center for Devices and RadiologicalHealth. MDR data files. Available at:www.fda.gov/CDRH/mdrfile.html. Ac-cessed April 1, 2003.

4. U.S. Food and Drug Administration, Cen-ter for Devices and Radiological Health.Manufacturer and user facility device ex-perience database (MAUDE): file formatsfor FOI releasable data. Available at: www.fda.gov/cdrh/maude.html. Accessed April 1,2003.

5. Shellock FG. Reference manual for mag-netic resonance safety, implants, and de-vices: 2004 edition. Los Angeles, Calif:Biomedical Research Publishing, 2004.

6. Shellock FG. Pocket guide to MR proce-dures and metallic objects: update 2001.Philadelphia, Pa: Lippincott Williams &Wilkins, 2001.

7. Shellock FG. Magnetic resonance proce-dure: health effects and safety. Boca Ra-ton, Fla: CRC, 2001.

8. Schaefer DJ, Bourland JD, Nyenhuis JA.Review of patient safety in time-varyinggradient fields. J Magn Reson Imaging2000; 12:20–29.

9. Schenck JF. Safety of strong, static mag-netic fields. J Magn Reson Imaging 2000;12:2–19.

10. King KF, Schaefer DJ. Spiral scan periph-eral nerve stimulation. J Magn ResonImaging 2000; 12:164–170.

11. Shellock FG. Radiofrequency energy-in-duced heating during MR procedures: areview. J Magn Reson Imaging 2000; 12:30–36.

12. McJury M, Shellock FG. Auditory noiseassociated with MR procedures: a re-view. J Magn Reson Imaging 2000; 12:37–45.

13. Shellock FG, Kanal E. Policies, guide-lines, and recommendations for MR im-aging safety and patient management. JMagn Reson Imaging 1991; 1:97–101.

14. Kanal E, Shellock FG. Policies, guide-lines, and recommendations for MR im-aging safety and patient management:patient monitoring during MR examina-tions. J Magn Reson Imaging 1992;2:247–248.

15. Shellock FG, Kanal E. Guidelines andrecommendations for MR imaging safetyand patient management. III. Question-naire for screening patients before MRprocedures. The SMRI Safety Commit-tee. J Magn Reson Imaging 1994; 4:749–751.

16. International Electrotechnical Commis-sion. Medical electrical equipment: par-ticular requirements for the safety ofmagnetic resonance equipment for medi-cal diagnosis. International Standard IEC60601–2-33. Geneva, Switzerland: Inter-national Electrotechnical Commission,2002.

17. Kanal E, Barkovich J, Bell C, et al. Amer-ican College of Radiology white paperon MR safety. AJR Am J Roentgenol2002; 178:1335–1347.

18. Shellock, FG. MRI safety, bioeffects, andpatient management. Available at: www.MRIsafety.com.

19. Institute for Magnetic Resonance Safety,Education, and Research. IMRSER home.Available at: www.IMRSER.org.

20. Schenck JF, Dumoulin CL, RedingtonRW, Kressel HY, Elliott RT, McDougallIL. Human exposure to 4.0-tesla mag-netic fields in a whole-body scanner.Med Phys 1992; 19:1089–1098.

21. Kangarlu A, Burgess RE, Zhu H, et al.Cognitive, cardiac, and physiologicalsafety studies in ultra high field mag-netic resonance imaging. Magn ResonImaging 1999; 17:1407–1416.

22. Hong CZ, Shellock FG. Short-term expo-sure to a 1.5 tesla static magnetic fielddoes not effect somato-sensory evokedpotentials in man. Magn Reson Imaging1990; 8:65–69.

23. Shellock FG, Schaefer DJ, Crues JV. Ex-posure to a 1.5-T static magnetic fielddoes not alter body and skin tempera-tures in man. Magn Reson Med 1989;11:371–375.

24. Brody A, Embury S, Mentzer W, WinklerM, Gooding C. Preservation of sickle cellblood flow pattern(s) during MR imag-ing: an in vivo study. AJR Am J Roent-genol 1988; 151:139–141.

25. Tenforde T, Budinger T. Biological ef-fects and physical safety aspects of NMRimaging and in vivo spectroscopy. In:Thomas S, Dixon R, eds. NMR in medi-cine: instrumentation and clinical appli-cations. Medical monograph no. 14.New York, NY: American Association ofPhysicists in Medicine, 1986; 493–548.

26. Vogl T, Pasulus W, Fuchs A, Krafczyk S,Lissner J. Influence of magnetic reso-nance imaging on evoked potentials andnerve conduction velocities in humans.Invest Radiol 1991; 26:432–437.

27. Weiss J, Herrick RC, Taber KH, ContantC, Plishker GA. Bio-effects of high mag-netic fields: a study using a simple ani-mal model. Magn Reson Imaging 1992;10:689–694.

28. Yuh WT, Ehrhardt JC, Fisher DJ, ShieldsRK, Shellock FG. Phantom limb pain in-duced in amputee by strong magneticfields. J Magn Reson Imaging 1992;2:221–223.

29. Prasad N, Wright DA, Ford JJ, ThornbyJI. Safety of 4-T MR imaging: study ofeffects on developing frog embryos. Ra-diology 1990; 174:251–253.

30. Prasad N, Bushong S, Thronby J, BryanR, Hazelwood C, Harrell J. Effect of nu-clear magnetic resonance on chromo-somes of mouse bone marrow cells.Magn Reson Imaging 1984; 2:37–39.

31. Geard C, Osmak R, Hall E, Simon H,Maudsley A, Hilal S. Magnetic resonanceand ionizing radiation: a comparative


Ra

dio

logy

evaluation in vitro of oncogenic andgenotoxic potential. Radiology 1984;152:199–202.

32. Cooke P, Morris P. The effects of NMRexposure of living organism. II. A ge-netic study of human lymphocytes. Br JRadiol 1981; 54:622–625.

33. Schwartz J, Crooks L. NMR imaging pro-duces no observable mutations or cyto-toxicity in mammalian cells. AJR Am JRoentgenol 1982; 139:583–585.

34. Tyndall DA, Sulik KK. Effects of mag-netic resonance imaging on eye devel-opment in the C57BL/6J mouse. Teratol-ogy 1991; 43:263–275.

35. Heinrichs WL, Fong P, Flannery M, et al.Midgestational exposure of pregnantBALB/c mice to magnetic resonance im-aging. Magn Reson Imaging 1988;6:305–313.

36. McRobbie D, Foster MA. Pulsed mag-netic field exposure during pregnancyand implications for NMR foetal imag-ing: a study with mice. Magn Reson Im-aging 1985; 3:231–234.

37. Brockway J, Bream P. Does memory lossoccur after MR imaging? J Magn ResonImaging 1992; 2:721–728.

38. Besson J, Foreman E, Eastwood L, SmithF, Ashcroft G. Cognitive evaluation fol-lowing NMR imaging of the brain.J Neurol Neurosurg Psychiatry 1984; 47:314–316.

39. Nyenhuis JA, Bourland J, Kildishev AV,Schaefer DJ. Health effects and safety ofintense gradient fields. In: Shellock FG,ed. Magnetic resonance procedures:health effects and safety. Boca Raton,Fla: CRC, 2001; 31–54.

40. Bourland JD, Nyenhuis JA, SchaeferDJ. Physiologic effects of intense MRIgradient fields. Neuroimaging Clin NAm 1999; 9:363–377.

41. Cohen MS, Weisskoff R, Kantor H. Sen-sory stimulation by time varying mag-netic fields. Magn Reson Med 1990; 14:409–414.

42. Budinger TF, Fischer H, Hentshel D, Rein-flder HE, Schmitt F. Physiological effectsof fast oscillating magnetic field gradi-ents. J Comput Assist Tomogr 1991; 15:909–914.

43. Mansfield P, Harvey PR. Limits to neuralstimulation in echo-planar imaging.Magn Reson Med 1993; 29:746–758.

44. Irnich W, Schmitt F. Magnetostimula-tion in MRI. Magn Reson Med 1995; 33:619–623.

45. Ehrhardt JC, Lin CS, Magnotta VA,Fisher DJ, Yuh WT. Peripheral nervestimulation in a whole-body echo-pla-nar imaging system. J Magn Reson Im-aging 1997; 7:405–409.

46. Abart J, Eberhardt K, Fischer H, et al.Peripheral nerve stimulation by time-varying magnetic fields. J Comput AssistTomogr 1997; 21:532–538.

47. Ham CL, Engels JM, van de Weil GT,Machielsen A. Peripheral nerve stimula-tion during MRI: effects of high gradientamplitudes and switching rates. J MagnReson Imaging 1997; 7:933–937.

48. Nyenhuis JA, Bourland JD, Schaefer DJ.Analysis from a stimulation perspectiveof magnetic field patterns of MR gradi-ent coils. J Appl Phys 1997; 81:4314–4316.

49. McJury M. Acoustic noise and magnetic

resonance procedures. In: Shellock FG,ed. Magnetic resonance procedures:health effects and safety. Boca Raton,Fla: CRC, 2001; 115–138.

50. Brummett RE, Talbot JM, Charuhas P.Potential hearing loss resulting from MRimaging. Radiology 1988; 169:539–540.

51. Quirk ME, Letendre AJ, Ciottone RA,Lingley JF. Anxiety in patients undergo-ing MR imaging. Radiology 1989; 170:463–466.

52. Laurell GF. Combined effects of noiseand cisplatin: short- and long-term fol-low-up. Ann Otol Rhinol Laryngol 1992;101:969–976.

53. Philbin MK, Taber KH, Hayman LA. Pre-liminary report: changes in vital signs ofterm newborns during MR. AJNR Am JNeuroradiol 1996; 17:1033–1036.

54. Counter SA, Olofsson A, Borg E. Bjelke B,Haggstrom A, Grahn HF. Analysis ofmagnetic resonance imaging acousticnoise generated by a 4.7 T experimentalsystem. Acta Otolaryngol 2000; 120:739–743.

55. Price DL, De Wilde JP, Papadaki AM,Curran JS, Kitney RI. Investigation ofacoustic noise on 15 MRI scanners from0.2 T to 3 T. J Magn Reson Imaging2001; 13:288–293.

56. Shellock FG, Ziarati M, Atkinson D,Chen DY. Determination of gradientmagnetic field-induced acoustic noiseassociated with the use of echo planarand three-dimensional fast spin echotechniques. J Magn Reson Imaging1998; 8:1154–1157.

57. Strainer JC, Ulmer JL, Yetkin FZ, Haugh-ton VM, Daniels DL, Millen SJ. Func-tional MR of the primary auditory cor-tex: an analysis of pure tone activationand tone discrimination. AJNR Am JNeuroradiol 1997; 18:601–610.

58. Ulmer JL, Bharat BB, Leighton PM, et al.Acoustic echo planar scanner noise andpure tone hearing thresholds: the effectsof sequence repetition times and acous-tic noise rates. J Comput Assist Tomogr1998; 22:480–486.

59. Bandettini PA, Jesmanowicz A, Van Ky-len J, Birn RA, Hyde J. Functional MRI ofbrain activation induced by scanneracoustic noise. Magn Reson Med 1998;39:410–416.

60. Goldman AM, Gossman WE, Fried-lander PC. Reduction of sound levelswith antinoise in MR imaging. Radiol-ogy 1989; 173:549–550.

61. Hurwitz R, Lane SR, Bell RA, Brant-Zawadzki MN. Acoustic analysis of gra-dient-coil noise in MR imaging. Radiol-ogy 1989; 173:545–548.

62. Counter SA, Olofsson A, Grahn, Borg E.MRI acoustic noise: sound pressure andfrequency analysis. J Magn Reson Imag-ing 1997; 7:606–611.

63. Hedeen RA, Edelstein WA. Characteris-tics and prediction of gradient acousticnoise in MR imagers. Magn Reson Med1997; 37:7–10.

64. Cho ZH, Park SH, Kim JH, et al. Analysisof acoustic noise in MRI. Magn ResonImaging 1997; 15:815–822.

65. Miller LE, Keller AM. Regulation of oc-cupational noise. In: Harris CM, ed.Handbook of noise control. New York,NY: McGraw-Hill, 1979; 1–16.

66. Melnick W. Hearing loss from noise ex-

posure. In: Harris CM, ed. Handbook ofnoise control. New York, NY: McGraw-Hill, 1979; 2.

67. Schaefer DJ. Safety aspects of radio-fre-quency power deposition in magneticresonance. Magn Reson Imaging Clin NAm 1998; 6:775–789.

68. Morvan D, Leroy-Willig A, Jehenson P,Cuenod CA, Syrota A. Temperaturechanges induced in human muscle byradiofrequency H-1 coupling: measure-ment with an MR imaging diffusiontechnique. Radiology 1992; 185:871–874.

69. Shellock FG, Crues JV. Temperature,heart rate, and blood pressure changesassociated with clinical MR imaging at1.5 T. Radiology 1987; 163:259–262.

70. Shellock FG, Crues JV. Temperaturechanges caused by clinical MR imagingof the brain at 1.5 tesla using a head coil.AJNR Am J Neuroradiol 1988; 9:287–291.

71. Shellock FG, Rothman B, Sarti D. Heat-ing of the scrotum by high-field-strength MR imaging. AJR Am J Roent-genol 1990; 154:1229–1232.

72. Shellock FG, Schatz CJ. Increases in cor-neal temperature caused by MR imagingof the eye with a dedicated local coil.Radiology 1992; 185:697–699.

73. Shellock FG, Schaefer DJ, Crues JV. Al-terations in body and skin temperaturescaused by MR imaging: is the recom-mended exposure for radiofrequency ra-diation too conservative? Br J Radiol1989; 62:904–909.

74. Kangarlu A, Shellock FG, Chakeres D.8.0-tesla MR system: temperature changesassociated with radiofrequency-inducedheating of a head phantom. J Magn ResonImaging 2003; 17:220–226.

75. Shellock FG, Schaefer DJ, Kanal E. Phys-iologic responses to MR imaging per-formed at an SAR level of 6.0 W/kg. Ra-diology 1994; 192:865–868.

76. Schaefer DJ. Health effects and safety ofradiofrequency power deposition associ-ated with magnetic resonance proce-dures. In: Shellock FG, ed. Magnetic res-onance procedures: health effects andsafety. Boca Raton, Fla: CRC, 2001; 55–74.

77. Shellock FG, Schaefer DJ. Radiofre-quency energy-induced heating duringmagnetic resonance procedures: labora-tory and clinical experiences. In: Shel-lock FG, ed. Magnetic resonance proce-dures: health effects and safety. BocaRaton, Fla: CRC, 2001; 75–96.

78. Shuman WP, Haynor DR, Guy AW,Wesby GE, Schaefer DJ, Moss AA. Super-ficial and deep-tissue increases in anes-thetized dogs during exposure to highspecific absorption rates in a 1.5-T MRimager. Radiology 1988; 167:551–554.

79. Gordon CJ. Normalizing the thermal ef-fects of radiofrequency radiation: bodymass versus total body surface area. Bio-electromagnetics 1987; 8:111–118.

80. National Council on Radiation Protec-tion and Measurements. Biological ef-fects and exposure criteria for radiofre-quency electromagnetic fields. Reportno. 86. Bethesda, Md: National Councilon Radiation Protection and Measure-ments, 1986.

81. Rowell LB. Cardiovascular aspects of hu-


Ra

dio

logy

man thermoregulation. Circ Res 1983;52:367–379.

82. Drinkwater BL, Horvath SM. Heat toler-ance and aging. Med Sci Sports 1979;11:49–55.

83. Kenney WL. Physiological correlates ofheat intolerance. Sports Med 1985; 2:279–286.

84. Barany FR. Abnormal vascular reactionin diabetes mellitus. Acta Med ScandSuppl 1955; 304:556–624.

85. Buskirk ER, Lundegren H, Magnusson L.Heat acclimation patterns in obese andlean individuals. Ann N Y Acad Sci 1965;131:637–653.

86. Jauchem JR. Effects of drugs on thermalresponses to microwaves. Gen Pharma-col 1985; 16:307–310.

87. Schaefer DJ, Barber BJ, Gordon CJ. Ther-mal effects of magnetic resonance imag-ing (abstr). In: Book of abstracts: Societyof Magnetic Resonance in Medicine1985. Berkeley, Calif: Society of Mag-netic Resonance in Medicine, 1985; 925.

88. Sawyer-Glover A, Shellock FG. Pre-mag-netic resonance procedure screening. In:Shellock FG, ed. Magnetic resonanceprocedures: health effects and safety.Boca Raton, Fla: CRC, 2001; 271–326.

89. Sawyer-Glover A, Shellock FG. Pre-MRIprocedure screening: recommendationsand safety considerations for biomedicalimplants and devices. J Magn Reson Im-aging 2000; 12:92–106.

90. Kelly WM, Paglen PG, Pearson JA, SanDiego AG, Soloman MA. Ferromag-netism of intraocular foreign bodycauses unilateral blindness after MRstudy. AJNR Am J Neuroradiol 1986;7:243–245.

91. Otto PM, Otto RA, Virapongse C, et al.Screening test for detection of metallicforeign objects in the orbit before mag-netic resonance imaging. Invest Radiol1992; 27:308–311.

92. Jarvik JG, Ramsey JG. Radiographicscreening for orbital foreign bodies priorto MR imaging: is it worth it? AJNR Am JNeuroradiol 2000; 21:245–246.

93. Seidenwurm DJ, McDonnell CH 3rd,Raghavan N, Breslau J. Cost utility anal-ysis of radiographic screening for an or-bital foreign body before MR imaging.AJNR Am J Neuroradiol 2000; 21:426–433.

94. Dempsey MF, Condon B, Hadley DM.Investigation of the factors responsiblefor burns during MRI. J Magn Reson Im-aging 2001; 13:627–631.

95. A new MRI complication? Health De-vices Alert, May 27, 1988; 1.

96. Thermal injuries and patient monitor-ing during MRI studies. Health Devices1991; 20:362–363.

97. Jones S, Jaffe W, Alvi R. Burns associatedwith electrocardiographic monitoringduring magnetic resonance imaging.Burns 1996; 22:420–421.

98. Keens SJ, Laurence AS. Burns caused byECG monitoring during MRI imaging.Anaesthesia 1996; 51:1188–1189.

99. Kugel H, Bremer C, Puschel M, et al.Hazardous situation in the MR bore: in-duction in ECG leads causes fire. EurRadiol 2003; 13:690–694.

100. Nakamura T, Fukuda K, Hayakawa K, etal. Mechanism of burn injury duringmagnetic resonance imaging (MRI):

simple loops can induce heat injury.Front Med Biol Eng 2001; 11:117–129.

101. Shellock FG, Slimp G. Severe burn of thefinger caused by using a pulse oximeterduring MR imaging (letter). AJR Am JRoentgenol 1989; 153:1105.

102. Smith CD, Nyenhuis JA, Kildishev AV.Health effects of induced electrical cur-rents: implications for implants. In:Shellock FG, ed. Magnetic resonanceprocedures: health effects and safety.Boca Raton, Fla: CRC, 2001; 393–413.

103. Shellock FG. MR safety update 2002: im-plants and devices. J Magn Reson Imag-ing 2002; 16:485–496.

104. Chou CK, McDougall JA, Can KW. Ab-sence of radiofrequency heating fromauditory implants during magnetic res-onance imaging. Bioelectromagnetics1995; 16:307–316.

105. Nyenhuis JA, Kildishev AV, Foster KS,Graber G, Athey W. Heating near im-planted medical devices by the MRI RF-magnetic field. IEEE Transactions onMagnetics 1999; 35:4133–4135.

106. Rezai AR, Finelli D, Nyenhuis JA, et al.Neurostimulator for deep brain stimula-tion: ex vivo evaluation of MRI-relatedheating at 1.5-tesla. J Magn Reson Imag-ing 2002; 15:241–250.

107. Finelli DA, Rezai AR, Ruggieri PM, et al.MR imaging-related heating of deepbrain stimulation electrodes: in vitrostudy. AJNR Am J Neuroradiol 2002; 23:1795–1802.

108. Shellock FG, Hatfield M, Simon BJ, et al.Implantable spinal fusion stimulator:assessment of MRI safety. J Magn ResonImaging 2000; 12:214–223.

109. Knopp MV, Essig M, Debus J, Zabel HJ,van Kaick G. Unusual burns of the lowerextremities caused by a closed conduct-ing loop in a patient at MR imaging.Radiology 1996; 200:572–575.

110. Becker H. The use of intradermal tattooto enhance the final result of nipple-areola reconstruction. Plast ReconstrSurg 1986; 77:673–676.

111. Halder RM, Pham HN, Hreadon JY,Johnson HA. Micropigmentation for thetreatment of vitiligo. J Dermatol SurgOncol 1989; 15:1092–1098.

112. Carr JJ. Danger in performing MR imag-ing on women who have tattooed eye-liner or similar types of permanent cos-metic injections. AJR Am J Roentgenol1995; 165:1546–1547.

113. Jackson JG, Acker H. Permanent eyelinerand MR imaging (letter). AJR Am JRoentgenol 1987; 149:1080.

114. Lund G, Wirtschafter JD, Nelson JD,Williams PA. Tattooing of eyelids: mag-netic resonance imaging artifacts. Oph-thalmic Surg 1986; 17:550–553.

115. Tope WD, Shellock FG. Magnetic reso-nance imaging and permanent cosmet-ics (tattoos): survey of complicationsand adverse events. J Magn Reson Imag-ing 2002; 15:180–184.

116. U.S. Food and Drug Administration,Center for Food Safety and Applied Nu-trition. Office of Cosmetics and Colorsfact sheet: tattoos and permanentmakeup. Rockville, Md: U.S. Food andDrug Administration, 2000.

117. Vahlensieck M. Tattoo-related cutane-ous inflammation (burn grade I) in a

mid-field MR scanner (letter). Eur Radiol2000; 10:197.

118. Wagle WA, Smith M. Tattoo-inducedskin burn during MR imaging (letter).AJR Am J Roentgenol 2000; 174:1795.

119. Kreidstein ML, Giguere D, Freiberg A.MRI interaction with tattoo pigments:case report, pathophysiology, and man-agement. Plast Reconstr Surg 1997; 99:1717–1720.

120. Kanal E, Shellock FG. MRI interactionwith tattoo pigments (letter). Plast Re-constr Surg 1998; 101:1150–1151.

121. Weiss RA, Saint-Louis LA, Haik BG, Mc-Cord CD, Taveras JL. Mascara and eye-lining tattoos: MRI artifacts. Ann Oph-thalmol 1989; 21:129–131.

122. Colletti PM. Magnetic resonance proce-dures and pregnancy. In: Shellock FG,ed. Magnetic resonance procedures:health effects and safety. Boca Raton,Fla: CRC, 2001; 149–182.

123. McCarthy SM, Filly RA, Stark DD, et al.Obstetrical magnetic resonance imag-ing: fetal anatomy. Radiology 1985; 154:427–432.

124. McCarthy SM, Stark DD, Filly RA, CallenPW, Hricak H, Higgins CB. Obstetricalmagnetic resonance imaging: maternalanatomy. Radiology 1985; 154:421–425.

125. Tempany CM, Fielding JR. Female pel-vis. In: Edelman RR, Hesselink JR, Zlat-kin MB, eds. Clinical magnetic reso-nance imaging. 2nd ed. Philadelphia,Pa: Saunders, 1996; 1432–1466.

126. Tyndall DA. MRI effects on the teratoge-nicity of X-irradiation in the C57BL/6Jmouse. Magn Reson Imaging 1990;8:423–433.

127. Yip YP, Capriotti C, Yip JW. Effects ofMR exposure on axonal outgrowth inthe sympathetic nervous system of thechick. J Magn Reson Imaging 1995;5:457–462.

128. Shellock FG. Biomedical implants anddevices: assessment of magnetic field in-teractions with a 3-tesla MR system. JMagn Reson Imaging 2002; 16:721–732.

129. Shellock FG, Tkach JA, Ruggieri PM,Masaryk T, Rasmussen P. Aneurysmclips: evaluation of magnetic field inter-actions using “long-bore” and “short-bore” 3-tesla MR systems. AJNR Am JNeuroradiol 2003; 24:463–471.

130. Shellock FG, Tkach JA, Ruggieri PM,Masaryk TJ. Cardiac pacemakers, ICDs,and loop recorder: evaluation of trans-lational attraction using conventional(“long-bore”) and “short-bore” 1.5- and3-tesla MR systems. J Cardiovasc MagnReson 2003; 5:387–397.

131. Kangarlu A, Shellock FG. Aneurysmclips: evaluation of magnetic field inter-actions with an 8.0 T MR system. J MagnReson Imaging 2000; 12:107–111.

132. Yuh WT, Hanigan MT, Nerad JA, et al.Extrusion of a magnetic eye implant af-ter MR examination: a potential hazardto the enucleated eye. J Magn Reson Im-aging 1991; 1:711–713.

133. Fagan LL, Shellock FG, Brenner RJ, Roth-man B. Ex vivo evaluation of ferromag-netism, heating, and artifacts of breasttissue expanders exposed to a 1.5 T MRsystem. J Magn Reson Imaging 1995;5:614–616.

134. Shellock FG. MR imaging and cervicalfixation devices: evaluation of ferromag-


Ra

dio

logy

netism, heating, and artifacts at 1.5 tesla.Magn Reson Imaging 1996; 14:1093–1098.

135. Shellock FG, Shellock VJ. Evaluation ofcranial flap fixation clamps for compat-ibility with MR imaging. Radiology1998; 207:822–825.

136. Shellock FG, Shellock VJ. Cardiovascularcatheters and accessories: ex vivo testingof ferromagnetism, heating, and arti-facts associated with MRI. J Magn ResonImaging 1998; 8:1338–1342.

137. Becker RL, Norfray JF, Teitelbaum GP, etal. MR imaging in patients with intra-cranial aneurysm clips. AJNR Am J Neu-roradiol 1988; 9:885–889.

138. Brothers MF, Fox AJ, Lee DH, Pelz DM,Deveikis JP. MR imaging after surgeryfor vertebrobasilar aneurysm. AJNRAm J Neuroradiol 1990; 11:149–161.

139. Brown MA, Carden JA, Coleman RE, etal. Magnetic field effects on surgical li-gation clips. Magn Reson Imaging 1987;5:443–453.

140. Burtscher IM, Owman T, Romner B,Stahlberg F, Holtas S. Aneurysm clip MRartifacts: titanium versus stainless steeland influence of imaging parameters.Acta Radiol 1998; 39:70–76.

141. Kanal E, Shellock FG. Aneurysm clips:effects of long-term and multiple expo-sures to a 1.5-T MR system. Radiology1999; 210:563–565.

142. Kanal E, Shellock FG, Lewin JS. Aneu-rysm clip testing for ferromagneticproperties: clip variability issues. Radiol-ogy 1996; 200:576–578.

143. Klucznik RP, Carrier DA, Pyka R, HaidRW. Placement of a ferromagnetic intra-cerebral aneurysm clip in a magneticfield with a fatal outcome. Radiology1993; 187:855–856.

144. New PF, Rosen BR, Brady TJ, et al. Po-tential hazards and artifacts of ferromag-netic and nonferromagnetic surgicaland dental materials and devices in nu-clear magnetic resonance imaging. Radi-ology 1983; 147:139–148.

145. Pride GL, Kowal J, Mendelsohn DB,Chason DP, Fleckenstein JL. Safety ofMR scanning in patients with nonferro-magnetic aneurysm clips. J Magn ResonImaging 2000; 12:198–200.

146. Shellock FG, Kanal E. Yasargil clips: eval-uation of interactions with a 1.5-T MRsystem. Radiology 1998; 207:587–591.

147. Romner B, Olsson M, Ljunggren B, et al.Magnetic resonance imaging and aneu-rysm clips. J Neurosurg 1989; 70:426–431.

148. Kato Y, Sano H, Katada K, et al. Effects ofnew titanium cerebral aneurysm clipson MRI and CT images. Minim InvasiveNeurosurg 1996; 39:82–85.

149. Lawton MT, Heiserman JE, PrendergastVC, Zabramski JM, Spetzler RF. Tita-nium aneurysm clips. III. Clinical appli-cation in 16 patients with subarachnoidhemorrhage. Neurosurgery 1996; 38:1170–1175.

150. Piepgras A, Guckel F, Weik T, SchmiedekP. Titanium aneurysm clips and theiradvantages in diagnostic imaging. Ra-diologe 1995; 35:830–833.

151. Wichmann W, Von Ammon K, Fink U,Weik T, Yasargil GM. Aneurysm clipsmade of titanium: characteristics and ar-

tifacts in MR. AJNR Am J Neuroradiol1997; 18:939–944.

152. Shellock FG, Shellock VJ. MR-compati-bility evaluation of the Spetzler tita-nium aneurysm clip. Radiology 1998;206:838–841.

153. Soulen RL. Reply (letter). Radiology1986; 158:279.

154. Soulen RL, Budinger TF, Higgins CB.Magnetic resonance imaging of pros-thetic heart valves. Radiology 1985; 154:705–707.

155. Edwards MB, Taylor KM, Shellock FG.Prosthetic heart valves: evaluation ofmagnetic field interactions, heating,and artifacts at 1.5-tesla. J Magn ResonImaging 2000; 12:363–369.

156. Frank H, Buxbaum P, Huber L, et al. Invitro behavior of mechanical heartvalves in 1.5 T superconducting magnet.Eur J Radiol 1992; 2:555–558.

157. Pruefer D, Kalden P, Schreiber W, et al.In vitro investigation of prosthetic heartvalves in magnetic resonance imaging:evaluation of potential hazards. J HeartValve Dis 2001; 10:410–414.

158. Shellock FG. Prosthetic heart valves andannuloplasty rings: assessment of mag-netic field interactions, heating, and ar-tifacts at 1.5 tesla. J Cardiovasc MagnReson 2001; 3:317–324.

159. Teitelbaum GP, Bradley WG, Klein BD.MR imaging artifacts, ferromagnetism,and magnetic torque of intravascular fil-ters, stents, and coils. Radiology 1988;166:657–664.

160. Shellock FG, Shellock VJ. Stents: evalua-tion of MRI safety. AJR Am J Roentgenol1999; 173:543–546.

161. Hennemeyer CT, Wicklow K, FeinbergDA, Derdeyn CP. In vitro evaluation ofplatinum Guglielmi detachable coils at3-T with a porcine model: safety issuesand artifacts. Radiology 2001; 219:732–737.

162. Hug J, Nagel E, Bornstedt A, Schacken-burg B, Oswald H, Fleck E. Coronary ar-terial stents: safety and artifacts duringMR imaging. Radiology 2000; 216:781–787.

163. Kiproff PM, Deeb DL, Contractor FM,Khoury MB. Magnetic resonance charac-teristics of the LGM vena cava filter:technical note. Cardiovasc Intervent Ra-diol 1991; 14:254–255.

164. Manke C, Nitz WR, Djavidani B, et al.MR imaging–guided stent placement iniliac arterial stenoses: a feasibility study.Radiology 2001; 219:527–534.

165. Marshall MW, Teitelbaum GP, Kim HS,et al. Ferromagnetism and magnetic res-onance artifacts of platinum emboliza-tion microcoils. Cardiovasc InterventRadiol 1991; 14:163–166.

166. Rutledge JM, Vick GW, Mullins CE,Grifka RG. Safety of magnetic resonanceimmediately following Palmaz stent im-plant: a report of three cases. CatheterCardiovasc Interv 2001; 53:519–523.

167. Buecker A, Neuerburg JM, Adam GB, etal. Real-time MR fluoroscopy for MR-guided iliac artery stent placement. JMagn Reson Imaging 2000; 12:616–622.

168. Spuentrup E, Ruebben A, Schaeffter T,Manning WJ, Gunther RW, Buecker A.Magnetic resonance-guided coronary ar-

tery stent placement in a swine model.Circulation 2002; 105:874–879.

169. Taal BG, Muller SH, Boot H, Koop W.Potential risks and artifacts of magneticresonance imaging of self-expandableesophageal stents. Gastrointest Endosc1997; 46:424–429.

170. Shellock FG. New metallic implant usedfor permanent female contraception:evaluation of MR safety. AJR Am JRoentgenol 2002; 178:1513–1516.

171. Erlebacher JA, Cahill PT, Pannizzo F,Knowles RJ. Effect of magnetic reso-nance imaging on DDD pacemakers.Am J Cardiol 1986; 57:437–440.

172. Hayes DL, Holmes DR, Gray JE. Effect of1.5 tesla nuclear magnetic resonanceimaging scanner on implanted perma-nent pacemakers. J Am Coll Cardiol1987; 10:782–786.

173. Holmes DR Jr, Hayes DL, Gray JE, Meri-deth J. The effects of magnetic reso-nance imaging on implantable pulsegenerators. Pacing Clin Electrophysiol1986; 9:360–370.

174. Zimmermann BH, Faul DD. Artifactsand hazards in NMR imaging due tometal implants and cardiac pacemakers.Diagn Imaging Clin Med 1984; 53:53–56.

175. Achenbach S, Moshage W, Diem B,Bieberle T, Schibgilla V, Bachman K. Ef-fects of magnetic resonance imaging oncardiac pacemakers and electrodes. AmHeart J 1997; 134:467–473.

176. Peden CJ, Collins AG, Butson PC, Whit-man J, Young I. Induction of microcur-rents in critically ill patients in magneticresonance systems. Crit Care Med 1993;21:1923–1928.

177. Fontaine JM, Mohamed FB, Gottlieb C,Callans DJ, Marchlinski FE. Rapid ven-tricular pacing in a pacemaker patientundergoing magnetic resonance imag-ing. Pacing Clin Electrophysiol 1998;21:1336–1339.

178. Fetter J, Aram G, Holmes DR, Gray JE,Hayes DL. The effects of nuclear mag-netic resonance imagers on external andimplantable pulse generators. PacingClin Electrophysiol 1984; 7:720–727.

179. Vahlhaus C, Sommer T, Lewalter T, et al.Interference with cardiac pacemakers bymagnetic resonance imaging: are thereirreversible changes at 0.5 tesla? PacingClin Electrophysiol 2001; 24(4 pt 1):489–495.

180. Gimbel JR, Johnson D, Levine PA,Wilkoff BL. Safe performance of mag-netic resonance imaging on five patientswith permanent cardiac pacemakers.Pacing Clin Electrophysiol 1996; 19:913–919.

181. Duru F, Luechinger R, Scheidegger MB,Luscher TF, Boesiger P, Candinas R. Pac-ing in magnetic resonance imaging en-vironment: clinical and technical con-siderations on compatibility. Eur Heart J2001; 22:113–124.

182. Pavlicek W, Geisinger M, Castle L, et al.The effects of nuclear magnetic reso-nance on patients with cardiac pace-makers. Radiology 1983; 147:149–153.

183. Luechinger R, Duru F, Scheidegger MB,Boesiger P, Candinas R. Force andtorque effects of a 1.5 tesla MRI scanneron cardiac pacemakers and ICDs. PacingClin Electrophysiol 2001; 24:199–205.


Ra

dio

logy

184. Alagona P, Toole JC, Maniscalco BS,Glover MU, Abernathy GT, Prida XE.Nuclear magnetic resonance imaging ina patient with a DDD pacemaker (letter).Pacing Clin Electrophysiol 1989; 12:619.

185. Shellock FG, O’Neil M, Ivans V, et al.Cardiac pacemakers and implantablecardioverter defibrillators are unaffectedby operation of extremity MR imagingsystem. AJR Am J Roentgenol 1999; 172:165–170.

186. Inbar S, Larson J, Burt T, Mafee M, EzriMD. Case report: nuclear magnetic res-onance imaging in a patient with apacemaker. Am J Med Sci 1993; 305:174–175.

187. Garcia-Bolao I, Albaladejo V, Benito A,Alegria E, Zubieta J. Magnetic resonanceimaging in a patient with a dual cham-ber pacemaker. Acta Cardiol 1998; 53:33–35.

188. Sommer T, Vahlhaus C, Lauck G, et al.MR imaging and cardiac pacemakers: in

vitro evaluation and in vivo studies in51 patients at 0.5 T. Radiology 2000;215:869–879.

189. Gimbel JR. Implantable pacemaker anddefibrillator safety in the MR environ-ment: new thoughts for the new millen-nium. In: Kanal E, ed. Syllabus: specialcross-specialty categorical course in di-agnostic radiology—practical MR safetyconsiderations for physicians, physi-cists, and technologists. Oak Brook, Ill:Radiological Society of North America,2001; 69–76.

190. Gimbel JR. Interference in implantedcardiac devices (letter). Pacing Clin Elec-trophysiol 2003; 26(4 pt 1):922–923.

191. Martin ET, Coman JA, Shellock FG, Pull-ing CC, Fair R, Jenkins K. Magnetic res-onance imaging and cardiac pacemakersafety at 1.5-tesla. J Am Coll Cardiol2004; 43:1315–1324.

192. Limousin P, Krack P, Pollak P, et al. Elec-trical stimulation of the subthalamic

nucleus in advanced Parkinson’s disease.N Engl J Med 1998; 339:1105–1111.

193. Rezai AR, Lozano AM, Crawley AP, et al.Thalamic stimulation and functionalmagnetic resonance imaging: localiza-tion of cortical and subcortical activa-tion with implanted electrodes. J Neuro-surg 1999; 90:583–590.

194. diPierro CG, Francel PC, Jackson TR, Ka-miryo T, Laws ER Jr. Optimizing accuracyin magnetic resonance imaging-guidedstereotaxis: a technique with validationbased on the anterior commissure-pos-terior commissure line. J Neurosurg1999; 90:94–100.

195. Rezai AR, Phillips M, Baker K, et al.Neurostimulation system used for deepbrain stimulation (DBS): MR safety is-sues and implications of failing to fol-low guidelines. Invest Radiol 2004; 39:300–303.


Ra

dio

logy

Reviews - Home | Diagnostic Radiology · 2018-04-05 · Johnson & Johnson, Medtronic, St Jude,...

Documents

Transcript of Reviews - Home | Diagnostic Radiology · 2018-04-05 · Johnson & Johnson, Medtronic, St Jude,...