1809558_vkRJSgG6YNXYheolDuv5em1pV

8/13/2019 1809558_vkRJSgG6YNXYheolDuv5em1pV

1/13

Magnetic ResonanceImaging ofBrain TumorLorena Tonarelli, B.Sc. (Hons), M.Sc.

INTRODUCTIONMagnetic resonance imaging (MRI) provides detailed

information about brain tumor anatomy, cellular struc-ture and vascular supply, making it an important tool forthe effective diagnosis, treatment and monitoring of thedisease. This article provides an overview of brain tumor, with a focus on gliomas, followed by a description of theprinciples of MRI signal and image generation. It thenreviews the most established MRI techniques for braintumor imaging, and their clinical utilities for differen-tial diagnosis, tumor grading, and response to treatmentassessment. The neurosurgical applications of MRI usedto maximize tumor resection while avoiding damage tohealthy brain tissue are also described.

OBJECTIVES

After reading this article, the reader will be able to:Define brain tumors and the role of MRI in theirdetection and treatmentExplain the principles of MRI signal and imagegenerationDiscuss advantages and limitations of MRItechniques used for brain tumorUnderstand the value of intraoperative MRI inimproving tumor resection and patient survival

BRAIN TUMORSBrain tumors are abnormal and uncontrolled prolif-

erations of cells. Some originate in the brain itself, in which case they are termed primary . Others spread tothis location from somewhere else in the body through

1.

2.

3.

4.

metastasis, and are termedsecondary . Primary braintumors do not spread to other body sites, and camalignant or benign. Secondary brain tumors are amalignant. Both types are potentially disabling anthreatening. Because the space inside the skull iited, their growth increases intracranial pressuremay cause edema, reduced blood flow, and dispment, with consequent degeneration, of healthy tthat controls vital functions.1,2

Brain tumors are, in fact, the second leading cof cancer-related deaths in children and young aAccording to the Central Brain Tumor Registry oUnited States (CBTRUS), there will be 64,530 newof primary brain and central nervous system tumornosed by the end of 2011. Overall, more than 60people currently live with the disease.3

Although the causes of brain tumors are unknowfew risk factors have been proposed. These includinjuries, hereditary syndromes, immunosuppressiolonged exposure to ionizing radiation, electromafields, cell phones, or chemicals like formaldehy vinyl chloride. None of these, however, is proven tally cause the disease.4

Symptoms of brain tumors include persistent headnausea and vomiting, eyesight, hearing and/or sproblems, walking and/or balance difficulties, peality changes, memory lapses, problems with cog

and concentration, and seizures.4

GLIOMASGliomas are the most common primary malignant

tumors. They originate from non-neuronal (glial) cells called astrocytes. The World Health Organiz(WHO) classifies them, from the least to the maggressive, into four grades5:

Grade Ipilocytic astrocytomaGrade IIdiffused astrocytomas

Grade IIIanaplastic astrocytomasGrade IVglioblastomasGliomas are typically associated with low su

This is due to a combination of factors, includingrelapse rates, which hinder successful treatment.6,7 Also,

CEwebsource.com is produced by Enterprises for Continuing Education Inc. (ECEI), PO Box 300, Brighton, MI 48116, (810) 229-3354. The views expressed in the journal are the authors and do not necessarily represent those of ECEI. Full and complete prescribing informa-tion should be reviewed regarding any product mentioned herein prior to use. ECEI hereby authorizes you to copy documents published by ECEI on the World Wide Web for non-commercial use only. In consideration of this authorization, you agree that any copy of these docu-

ments which you make shall retain all copyright and other proprietary notices contained herein.


2/13

MAGNETIC RESONANCE IMAGING OF BRAIN TUMOR

CEwebsource.com PAGE 2 OF 12 EXPIRES JULY 15, 2013

gliomas are highly invasive. They can infiltrate to sur-rounding and remote brain areas as single cells. So, thereis often no clear boundary between tumor and healthytissue, which makes surgical resection difficult.8

Infiltration may involve eloquent areas; that is, those with specif ic functions, such as language, vision, andmotor control. In this case, gliomas are referred to asinfiltrating. IV-grade gliomas include glioblastoma multi- forme , the most aggressive type of brain tumor. Its mediansurvival rate is just 12 months.8

MRI USE IN BRAIN TUMORMedical history, biopsywhereby a small amount of

brain tissue is excised and analyzed under the micro-scopeand imaging studies are all important to reach adiagnosis of brain tumor. Standard x-rays and computedtomography (CT) can initially be used in the diagnosticprocess. However, MRI is generally more useful becauseit provides more detailed information about tumor type,position and size. For this reason, MRI is the imagingstudy of choice for the diagnostic work up and, there-after, for surgery and monitoring treatment outcomes.9

MRI SIGNAL AND IMAGE GENERATION

Conventional MRI exploits three physical properties oftissue protons (Table 1) to generate signals that are imagedas areas of different contrast, which reflect the anatomyand physiology of the organ under investigation.10

Protons are positively-charged particles inside the

nucleus of elements atoms. Because we are mainly madeup of water, the most abundant element in our body ishydrogen, each atom of which has one proton. To under-

stand how the MRI signal is generated, we can ithis proton as a minute magnet bar that movesspinning top. This tiny magnet bar can be descra two-dimensional (2D) vectoralso called aspin vector with respect to a Cartesian coordinate system.11

Table 1. Protons Properties Contributing to MRISignal Generation* Proton density Number of hydrogen proto

per unit volume of tissue

T1 relaxation time Time needed to recover 63%the longitudinal magnetizat(see explanation in text)

T2 relaxation time Time needed for 63% of transverse magnetization tolost (see explanation in text)

*Data from Westbrook C. MRI at a Glance. Wiley-Blackwell; 20

In the absence of a magnetic field, the spin vrandomly orientated. However, when placed in astatic magnetic field (B0), such as that generated by Mscanners, the spin vector tends to align parallel w0As a result, its component along the y-axis (longitudinamagnetization) is different from zero but the one the x-axis (transverse magnetization) is zero, and no sigcan be produced.11,12

If a radio frequency (RF) pulse is applied, th vector absorbs energy from the latter and movealignment away from B0. (This phenomenon is caresonance and occurs only if the frequency of the aRF pulse matches the protons own natural osc

frequency.) Both transverse and longitudinal mations are now different from zero, and a signalgenerated (Figure 1).12

FIGURE 1. (A) Spin vector (blue arrow) aligned parallel with the main magnetic f ield (B 0 ). ransversemagnetization is zero, and no signal is generated. (B) Spin vector out of alignment, following theof an RF pulse. ransverse magnetization is different from zero, and a signal can be generated.


3/13



Switching off the RF pulse causes the spin vectorto release the energy absorbed from the RF pulseaprocess calledrelaxationand to realign with the mainmagnetic field (B0). When this occurs, the transversemagnetization decays, whereas the longitudinal mag-netization recovers. The time needed to recover 63% ofthe longitudinal magnetization is calledT1-relaxationtime . The time taken for 63% of the transverse mag-netization to be lost is calledT2-relaxation time . T1-and T2-relaxation times are specif ic to different typesof tissue, and increase as the magnets field strengthincreases.12

Both T1 and T2 are components of the MRI signal.A third component is proton density; that is, the numberof hydrogen protons per unit volume of the tissue beingimaged. All three contribute to contrast generation. But,because they do it simultaneously, the image would beconfusing and difficult to interpret. The problem isavoided by weighting the contrast toward one of the threesignals components. As a result, the following imagescan be obtained: T1-weighted, T2-weighted, and protondensity-weighted.10,12

In all three types of images, bright areas correspondto tissue with high signal intensity (i.e., large transverse magnetization) and are referred to as hyperintense, whereas dark areas correspond to low signal intensity(i.e., small transverse magnetization) and are referred toas hypointense.13

However, contrast is obtained in different waysFrom differences in the T1recovery times of tisues, in T1-weighted images.13

From differences in the T2decay times of tis-sues, in T2-weighted images.13

From differences in the proton density of sues, in proton-weighted images.13

In terms of reading and interpreting an MRIthis means that high-fat-content tissues appear asareas in T1-weighted images, whereas high-wattent tissues appear as dark areas. The opposite is T2-weighted images. Since most diseases are chized by increased water content in tissues, T2-wimages are particularly useful for pathological intions. In contrast, T1-weighted images are best ftomical studies, although they can be used for paif combined with contrast enhancement. Currenstandard contrast agent is gadolinium (Gd), duability to cross the blood-brain barrier (BBB).13

In proton-weighted images, bright areas indicatproton-density tissues, such as the cerebrospinaand dark areas indicate low-proton-density tissuas cortical bone. Proton-density images are geused to show anatomy and some pathology.13 (See Tab2, which provides an overview of the key feature T2-, and proton-weighted images at a glance.)

Image Type Contrast Tissue Appearance Best For

T1-weightedMainly due todifferences in T1 recoverytimes

High-fat -content tissues appear as bright areas of high signal intensity (hyperintense)High- water -content tissues appear asdark areas of low signal intensity (hypointense)

Anatomy and, if use with contrast enhanment, also pathology

T2-weightedMainly due todifferences in T2 decay times

High-fat -content tissues appear as darkareas of low signal intensity (hypointense)

High- water -content tissues appear asbright areas of high signal intensity (hyper-intense)

Pathology

Proton-density weighted

Mainly due todifferences inproton density

Low-proton-density tissues appear as darkareas of low signal intensity (hypointense)High-proton-density tissues appear asbright areas of high signal intensity (hyper-intense)

Anatomy, and somepathology

Table 2. Types, Characteristics and Use of Weighted Images*

*Data from Westbrook C. MRI at a Glance. Wiley-Blackwell; 2010.


4/13



BRAIN TUMOR MR IMAGINGPrimary malignant brain tumors, such as gliomas,

appear as hypointense, dark areas on gadolinium-enhanced T1-weighted MRI images, and as hyperintense,bright areas on T2-weighted images (Figures 2, 3). While interpretation of gadolinium-enhanced T1-

weighted images and T2-weighted images remainsthe mainstay of brain tumor diagnosis, this approach

has limitations. For example, it is sometimes dto differentiate new from old tumors, or tumornontumoral lesions, like ischemia. Grading, monof tumor progression, treatment response asseand detection of residual tumor after surgery mbe problematic. Techniques other than T1- an weighted MRI can help overcome these limit The following sections describe the most estaand widely available.14

FIGURE 2. (A) Axial 2-weighted image of low-grade glioma in the right hemisphere, showing thelesion as a bright area of high-signal intensity. (B) Te same glioma on a 1-weighted image with contenhancement, showing the lesion as a dark area of low-signal intensity. (Photo courtesy of Professor Wick and Professor Martin Bendszus, of the Neurooncology Department of Heidelberg University, G

FIGURE 3. (A) Axial 2-weighted image of IV-grade glioma (glioblastoma) in the right hemisphshowing the lesion as a bright area of high-signal intensity. (B) he same glioblastoma on a 1-wimage with contrast enhancement, showing the lesion as a dark area of low-signal intensity. (Phocourtesy of Professor Wolfgang Wick and Professor Martin Bendszus, of the Neurooncology Deof Heidelberg University, Germany.)


5/13



DIFFUSION-WEIGHTED MRI

In diffusion-weighted imaging (DWI), contrastreflects changes in the random, temperature-dependentmovement of water molecules. This movement istermeddiffusion.11,15,16

In biological tissues, structural barriers like cell mem-branes prevent water molecules from moving completelyat random. Therefore, the rate of water diffusion (i.e., themagnitude of water molecule movement) is expressed asapparent diffusion coefficient (ADC) value.11,15,16

Diffusion-weighted images are acquired by superim-posing, for a short time, strong field gradients, called gradient pulses , on the main magnetic field (B0). Thisresults in image sensitization to water in the directionof the gradients, and signal attenuation along the axis to

which these are applied.11,17,18

Sensitization to water motion is defined by a param-eter called gradient factor or b value (sec/mm2). Thehigher the diffusion in the tissue under study thegreater the signal attenuation and the darker the corre-spondent area on the image. Diffusion data can also bepresented as mathematical maps of the apparent dif-fusion coefficient of water obtained from DW images.In this case, the higher the diffusion the smaller thesignal attenuation and the brighter the correspondentarea on the map (Table 3).11,17,18

Primary malignant brain tumors like gliomas usuallyshow lower cellularity and, consequently, greater diffu-

sion than normal tissue. Thus, they are dark on DWimages and bright on ADC maps.9Diffusion-weighted imaging is a valuable tool in the

follow-up of high-grade gliomas, particularly for dif-ferentiating treatment effects from tumor recurrenceor progression. This is because radiation and che-motherapythe standard treatments for brain tumortogether with surgerycause decreased tissue cellularityand, consequently, increased ADC, which is detectableas areas of higher signal intensity (i.e., brighter areas) within the tumor.19

In terms of diagnosis, DWI has advantages over otherMRI modalities, for it allows distinguishing malignant

from benign tumors, and tumoral from nontulesions. However, it has shortcomings. For examnoted earlier, gliomas in ADC maps normally sbright areas (high ADC). Yet, a small number oblastomas have shown as dark areas (low Astudies. This makes it difficult to distinguish theother types of brain tumors, like meningiomasalso have low ADC. It is therefore recommenassist with the diagnosis, to combine DWI withMRI modalities.14

PERFUSION-WEIGHTED MRI

Perfusion-weighted MRI measures perfusionblood flow) in tissues. It involves the use of an cular exogenous contrast agent, typically a bolus

of gadolinium administered during ultrafast T2 ations. The contrast agent causes reductions in T2 Thus a signal decay curve, called time intensit(TIC) can be generated after data acquisition.11-13 TICs for several images acquired during an

gadolinium injection are combined to produce a map of cerebral blood volume (CBV). On CBVareas of high perfusion, like in malignant brain are bright. Areas of low perfusion, such as in are dark.11-13 The microvessels of certain brain tumors, s

meningiomas and lymphomas, do not form a Ba result, gadolinium can leak into the interstitia

This does not occur in other types of brain tulike peripheral glioblastomas, whose microvessa BBB that prevents leakage of the contrast mAs a result, differences in time intensity curvbe observed, which help discriminate betweentypes and, therefore, assist with differential diaPerfusion imaging is also useful for glioma gespecially for ensuring surgical resection of thegrade area of heterogeneous brain tumors. In gCBV maps that demonstrate elevated blood vindicate the presence of a glioma, and grade-IVshould be suspected when values are similargreater than, that of the cortex.20-23

Table 3. Diffusion, Signal Intensity and Appearance in DW Images and ADC Maps*

DWImages

Increased diffusion greater signal attenuation lower signal intensity dark areaReduced diffusion smaller signal attenuation higher signal intensity bright area

ADCMaps

Increased diffusion smaller signal attenuation higher signal intensity bright areaReduced diffusion greater signal attenuation lower signal intensity dark area

*Data from Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E, Laval-Jeantet M. MR imaging of intravoxel incoherent mapplication to diffusion and perfusion in neurologic disorders.Radiology . 1986;161:401-7.


6/13



DIFFUSION-TENSOR MRI

As with DWI, diffusion-tensor MRI (DTI) detects

changes in diffusion of water molecules in tissues, vis-ible on diffusion-weighted images or ADC maps, butinvolves the use of strong multidirectional gradients toacquire additional data about the preferential directionand magnitude of the diffusion.24

In other words, DTI shows the degree to which water diffusion is facilitated in one direction, also calleddiffusion anisotropy . Mathematically, this is expressed bya quantity called fractional anisotropy (FA). FA measure-ments allow mapping, and establishing the integrity of,myelinated fiber tracts (white matter) that connect theprocessing centers of the cerebral cortex (gray matter)(Figure 4). Consequently, diffusion-tensor imaging can

be used to ascertain whether, and to which extent, thetumor has displaced, or otherwise damaged, eloquentareas. This, in turn, allows one to determine preopera-tively (i.e., before surgery) the tumors position respec-tive to such areas, and to predict the type and degree ofimpairment that may result from injuries caused intraop-eratively (i.e., during surgery).24-26

Diffusion-tensor imaging has other important uses.For example, one study of patients with meningioma orinfiltrating enhancing high-grade glioma has demon-strated that it is possible, with the appropriate choice ofb values, angular resolution, and signal-to-noise ratio, todistinguish one tumor type from the other by looking, on

ADC maps, at differences in FA measurements bareas of vasogenic edema and normal-appearin

matter surrounding the lesions.27

(Vasogenic edemathe accumulation of extracellular cerebrospinal fto disruption of the blood-brain barrier that tyforms around brain tumors.)

In the study, diffusion-tensor imaging was per with a 1.5 Tesla MR scanner (Signa; General EMedical Systems) using a standard head coil. Idata were acquired with a single-shot spin-echplanar sequence using the following protocol27:

Repetition time: 12,000 msecInversion time: 2,200 msec

Matrix size: 128 x 64b values: 0 and 1,000 sec/mm2 (six gradientdirections, one signal acquired)

DTI showed a difference in FA values in vasedema and normal-appearing white matter bemeningiomas and high-grade gliomas, which conto differential diagnosis.

Furthermore, a comparison among data acquir T1-, T2- and diffusion-tensor-weighted imaging ition-matched brain areas led researchers to suggDTI can help identify tumor infiltration that detected with conventional MRI.27

FIGURE 4. Human brain right dissected lateral view. A: Cerebral cortex (gray matter). B: White matteeral ventricle, frontal horn. 2: Lateral ventricle, central part. 3: Calcar avis. 4: Lateral ventricle, occipitCollateral trigone. 6: Collateral eminence. 7: Hippocampus. 8: Lateral ventricle, temporal horn. 9: Intesule. 10: Caudate Nucleus. (Photo courtesy of Dr. John A. Beal of the Department of Cellular Biologyof Louisiana State University Health Sciences Center, Shreveport, Louisiana.)


7/13



MAGNETIC RESONANCE SPECTROSCOPY

Magnetic resonance spectroscopy (MRS) is mainlyused for tissue metabolism studies and for differentiatingbetween tumor types. Data acquired with MRS are pre-sented as a spectrum, consisting of a graph showing therelation between signal intensity and proton resonancefrequency. MRS spectra vary with the chemical andmolecular structure of the tissue under study. They areused to evaluate frequency differences, or chemical shift,between tissues. The chemical shift is displayed on the x-axis in parts per million (ppm). On the y-axis, are arbi-trary units of signal intensity.12,13 Typically, MRS spectra show series of peaks corre-

sponding to metabolites found in a tissue. MRS datacan be used to generate metabolic maps that show where, in the brain, the peaks are located.12,13 Majorpeaks observed in 1.5 Tesla MRS of brain tumor includemyoinositol (3.6 ppm), creatine (two resonant peaks at3.0 and 3.9 ppm), choline (3.2 ppm) n-acetyl aspartate(NAA; 2.0 ppm), lactate (1.3 ppm) and lipid (one broadpeak from 0.9 to 1.5 ppm).14 The typical MRS spectrum of a glioma has low or

absent peaks of NAA (a marker of neuronal density and viability) and high peaks of choline (a marker of cellmembrane turnover). High-grade gliomas such as glio-blastomas also show lactate and lipid peaks.14

Since different types of brain tumors have similarspectra, MRS is not useful for differential diagnosis.However, it has proved to be a valuable tool for gliomagrading. For example, high choline/NAA peak height

ratios (above 1.5 ppm) have been associated in studies with high-grade lesions.28MRS is also used to:

More accurately target biopsies, by aiming atareas with high choline/NAA ratios29,30

Recognize tumor recurrence, as this shows as aprogressive increase of the choline/NAA ratio31

Detect radiation necrosis, which shows as acomplete absence, or progressive reduction overtime, of NAA and choline peaks, whereas lac-tate and lipid peaks are present29

To obtain useful data and avoid artifacts it is importantto choose an appropriate area of interest and size of voxel(i.e., the volume of patient tissue; defined as pixel areamultiplied slice thickness). Also, reliable MRS is gen-erally based on the comparison of several spectra, to seehow these change over time. Data acquisition requiresthe supervision of a trained spectroscopist.14

INTRAOPERATIVE MRIAs explained earlier, because gliomas spread by infil-

tration, it is difficult for the neurosurgeon to distinguishthem from normal brain tissue based on visual and tex-

tural inspection alone. Therefore, resecting thempletely is a major challenge. Yet, the extent of resection (also referred to ascraniotomy ) is crucial to s vival, regardless of the level of malignancy.32

Image-guided systems are available to assist trosurgeon during craniotomy, and allow for a moplete and accurate resection of the tumor. Knoneuronavigation systems, these include ultrasonocomputed tomography (CT), and MRI (computer-stereotaxis). However, conventional neuronavinvolves the use of image data acquired preope Therefore, it does not account for changes in bration and morphology that occur during the surgerchanges are collectively referred to asintraoperative braishift , and hinder the accurate localization and reof the tumor.33

Brain shift occurs at the start of the operationthe neurosurgeon opens the dura (i.e., the outermosurrounding the brain and spinal cord). This altpressure inside the skull causing the brain to movshift continues during the operation, because of loss of cerebrospinal fluid, swelling due to medand/or anesthetics, and of course collapse of the rcavity, as the neurosurgeon removes the tumor.33

Intraoperative magnetic resonance imaging ( whereby MRI scanning takes place immediatelyduring, and after surgery, can effectively compethese changes, providing consistently accurate gas to where the tumor is during the various stageoperation. This increases the chances of complettion, in turn improving survival. For this reason

is considered one of the most important develoin neurosurgery.34 The posit ive correlation between amount of

removed and survival has been widely demonIn one study, patients who had their glioma ontially removed were 4.9 times more likely to diethree years from the operation, than patients wthe tumor completely removed. Partial resection associated with a 1.4-fold increased risk of relapsthe same follow-up period.35

In the study, complete resection was achieved 0.5 Tesla intraoperative MRI system developedlaboration with General Electric Medical System

extent of tumor resection was monitored during by performing MRIat the initial positioning of the patient,after the opening of the dura, andat the end of the surgery.

Imaging data sets were acquired using fast spoidient (FSPGR), T1, T2, and gradient sequencesthickness was 1.5 to 2.5 mm.35

Each patients tumor area was measured ondimensional (3D) FSPGR images, as theseprecisely define the boundary between tumor atumoral tissue. If the SPGR images were not


8/13



for example, because of technical or motion artifacts, T2-weighted sequences of 5-mm-thick slices were usedinstead. In either case, the tumors volume was then cal-culated by adding all the voxels within the tumor areaon each slice. Volumetric calculations from SPGR or T2images obtained at the start of the surgery were com-pared with those taken at surgery completion. If thelatter showed no residual tumor volume, resection wasconsidered complete.35

In another study, total resection of low- and high-grade gliomas with no infiltration in eloquent areas was possible, with iMRI, in 100 percent of patients. Without iMR I, tota l resect ion was achieved onlyin 60 percent of patients. Total resection of gliomas with infiltration in eloquent areas was also achievedmore often with iMRI than without it (89.23 versus71.4 percent of patients).36

AWAKE CRANIOTOMY

Usually, patients whose preoperative MRI shows thatthe tumor has not spread to eloquent areas undergo sur-gery with general anesthesia. Those whose preoperativeMRI indicates that the tumor has infiltrated eloquentareas, undergo surgery with local anesthesia. The lattersurgical procedure is calledawake craniotomy .37

As noted earlier, resection of infiltrating gliomasposes special challenges. Recognizing and resecting onlytumoral tissue is extremely difficult. Therefore, there isa high risk of damaging eloquent areas, causing postop-erative neurologic deficits, such as loss of the ability to

talk, move, see, etc. To minimize this risk, the neuro-surgeon needs to be able to delineate the tumors mar-gins. This is achieved by performing what is known as cortical mapping .

In cortical mapping, patients are only locally anes-thetized, so they are awake and able to function duringthe operation. The procedure involves asking the patientto perform simple tasks, like naming objects, while theneurosurgeon concurrently stimulates their brain withan Ojemann cortical stimulator (Ojemann CorticalStimulator, Integra Radionics, Inc. Burlington, Mass). This is a battery-powered, hand-held device that allowssmall currents to be applied to localized areas of the brains

surface, temporarily deactivating them. If, during stimu-lation of a certain area, the patient performs a given tasknormally, the area being stimulated is safe to resect. Ifthe patient struggles to complete the task, the area beingstimulated must not be removed, even if tumor is present.Resection in that area would cause neurological damage.Cortical mapping is performed continuously during theoperation, allowing the neurosurgeon to define the limitsof safe tumor resection in real time.38,39

AWAKE CRANIOTOMY WITH MRI

The usefulness of awake craniotomy for p with glioma that is near to, or infilt rat ing, eloareas is enhanced when cortical mapping is per with the aid of MRI. Initially, this was achievethe earlier mentioned computer-guided sterotaone study, combining this technique with comapping achieved near-complete tumor resecdemonstrated by a greater than 90 percent rtion in T2 signal postoperativelyin 52 peof patients. There were no deaths during thgery, and although 74 percent of patients devneuronal deficits intraoperatively, approxi71 of these patients recovered within three mo40Imaging data were acquired just before surge48 hours after. The protocol included twosequences: a T2-weighted, spin echo sequenclong relaxation time, and a three-dimensionametric T1-weighted spin echo with short relatime and gadolinium enhancement. Both sequhad the following parameters40:

Field of view: 24 cmMatrix size: 256 x 256 pixelsSection thickness: 3 mm

However, because stereotaxis uses preopimages to guide the neurosurgeon, it cannot cosate for brain shift. Better outcomes can be obby performing awake craniotomy with intraop

MRI. In one study, this approach successfully atotal resection in eloquent areas with no disrupneurological function.41

Intraoperative MRI was performed with a 1.5unit, using the same sequences as in conventionscanning. A flexible surface coil was kept unpatients head for the entire duration of the surgsecond surface coil was placed on top of the Mclamp (used to keep the patients head fixed to gical table) only for scanning.41

A T2-weighted scan was performed soon aftetioning the patient, to identify potential artiDuring the surgery, scanning was performed on d

Routine T1-weighted images were used, befoafter contrast administration, for tumors showintrast enhancement. Fluid-attenuated inversion re(FLAIR) and axial T2-weighted imaging were unonenhancing tumors.41 To avoid the possibility that iMRI altered the

of cortical stimulation, surgery was performed inof the operating room outside the five-gauss lineoutside the static magnetic field generated by the Patients were transferred from the surgical sitescanner, and back, with a pivotal table. MRI imagloaded into the navigation system. The imaging was kept as short as possible.41


9/13



Patient transfer to and from the scanner required aboutthree to five minutes every time. Although this prolongedthe surgery by up to 60 minutes, the overall procedure was well tolerated by patients, the majority of whom(32 out of 34) said they would undergo a second awakecraniotomy with iMRI. Importantly, the use of intraop-erative MRI was not associated with adverse effects, suchas seizures.41

Besides eliminating the risk of MRI potentiallyaffecting cortical stimulation results, performing thesurgery outside the five-gauss line allows the use of fer-romagnetic tools and equipment. Early iMRI units weredesigned in a way that required both the patient and thesurgeon to be inside the magnet. Consequently, the sur-gery had to be performed with nonferromagnetic surgicalequipment (e.g., operative microscope, cortical stimu-lator, ultrasonic aspirator), which is more expensive thanthe standard equipment. In addition, costs increase withstronger magnetic fields. More recently developed iMRIscanners allow the neurosurgeon to operate outside themagnetic field.42

MRI scanning is not always performed in the oper-ating room. Some hospitals and other surgical centershave the MRI unit in an adjacent room, which allowsthe scanner to be used for outpatient imaging when notneeded for surgery.42

Performing surgery outside the magnetic field ispossible with both low-field (0.5 Tesla) and high-field(1.5 Tesla) systems. However, because of increasedmagnet strength and magnetic gradients, high-fieldiMRI yields significantly improved image quality and

signal-to-noise ratio. Furthermore, it is faster, allowsfor superior anatomical resolution, and enables theacquisition of sequences that are essential for gliomasurgery (e.g., T2-weighted images), all of which con-tribute to more complete tumor resection.42 In onestudy of 46 patients who underwent craniotomy, addi-tional tumor removal after high-field iMRI increasedthe extent of resection from 76 to 96 percent, andallowed neurosurgeons to achieve gross total resectionin 71 percent of cases.43

Positive results have also been obtained with ultra-high-field (3 Tesla) iMRI. In a recent studythefirst one to use an ultrahigh-field system for glioma

surgerythe number of total gross resections increasedby 32.3 percent with iMRI. Additionally, the intraoper-ative application of MRI made it possible to accuratelydiscriminate between T2-weighted changes in normaltissue and residual tumor.44

CONCLUSIONMRI is the preferred imaging study for brain tumor

diagnosis, providing detailed information on lesiontype, size and location. Although gadolinium-enhanced T1-weighted images and T2-weighted images are theMRI modalities of choice for the initial assessment,

their usefulness in identifying tumor types, guishing tumors from nontumoral lesions, and atreatment effects is limited. For this reason, thused in combination with other MRI techniquemost researched and employed of these techinclude diffusion-weighted imaging (DWI), peMRI, diffusion-tensor imaging (DTI) and maresonance spectroscopy (MRS).

Perhaps the most important application of maresonance is intraoperative MRI (iMRI), which alocalizing the tumor during surgery. Used with or cortical stimulation, iMRI can maximize tumortion while minimizing damage to healthy tissue, reducing the risk of neuronal deficit, and imppatient survival.

REFERENCESZeltzer PM.Brain Tumors: Leaving the Garden of Eden A Survival Guide to Diagnosis, Learning the Basics, GettingOrganized, and Finding Your Medical Team. Encino, Calif:Shilysca Press; 2004.Beatty J.Principles of Behavioral Neuroscience . Dubuque, IowaBrown & Benchmark Publishers; 1995.Central Brain Tumor Registry of the United States(CBTRUS). Fact Sheet. Available at: http://www.cbtruAccessed April 4, 2011.Viner J.Brain Tumors . University of California, San FrancDepartment of Neurosurgery. Available at: http://nursinmedicalcenter.org/education/classMaterial/34_2.pdf. AApril 6, 2011.

Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHclassification of tumours of the central nervous system Acta Neuropathol . 2007;114:97-109.Burger PC, Vogel FS, Green SB, Strike TA. Glioblastomultiforme and anaplastic astrocytoma. Pathologic criprognostic implications.Cancer . 1985;56:1106-11.Bellail AC, Hunter SB, Brat DJ, Tan C, Van Meir EG.Microregional extracellular matrix heterogeneity in brmodulates glioma cell invasion. Int J Biochem Cell Biol .2004;36:1046-69.Krex D, Klink B, Hartmann C, et al for the German GlNetwork. Long-term survival with glioblastoma multifBrain. 2007;130:2596-606.

McKhann II GM, ed.Clinical Neurosurgery, Vol. 53: APublication of the Congress of Neurological Surgeons (Clinic Neurosurgery). Boston, Mass: Wolters Kluwer, Lippincott Williams & Wilkins; 2006.Andreasen NC, ed.Brain Imaging: Applications in Psychiatry. Washington, DC: American Psychiatric Press; 1989.Debatin JF, McKinnon GC, eds.Ultrafast MRI: Techniques a Applications . Berlin, Germany: Springer-Verlag; 1998. Westbrook C, Roth CK, Talbot J, eds. MRI in Practice . 3rd edOxford, United Kingdom: Wiley-Blackwell; 2005. Westbrook C. MRI at a Glance. Hoboken, NJ: Wiley-Blackwell; 2010.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.


10/13



Young GS. Advanced MRI of adult brain tumors. Neurol Clin.2007;25:947-73.Leach MO. Spatially localized nuclear magnetic resonance. In: Webb ES, ed.The Physics of Medical Imaging. Bristol, UnitedKingdom: Institute of Physics Publishing; 1998.Hesselink JR. Diffusion-weighted imaging. Available at:

http://spinwarp.ucsd.edu/NeuroWeb/Text/br-710dwi.htm.Accessed April 4, 2011.Stehling MK, Turner R, Mansfield P. Echo-planar imaging:magnetic resonance imaging in a fraction of a second.Science .1991;254:43-50.Le Bihan D, Breton E, Lallemand D, Grenier P, CabanisE, Laval-Jeantet M. MR imaging of intravoxel incoherentmotions: application to diffusion and perfusion in neurologicdisorders.Radiology . 1986;161:401-7.Hein PA, Eskey CJ, Dunn JF, Hug EB. Diffusion-weightedimaging in the follow-up of treated high-grade gliomas:tumor recurrence versus radiation injury. Am J Neuroradiol .2004;25:201-9.Hartmann M, Heiland S, Harting I, et al. Distinguishing ofprimary cerebral lymphoma from high-grade glioma with per-fusion-weighted magnetic resonance imaging. Neurosci Lett .2003;338:119-22.Chaskis C, Stadnik T, Michotte A, Van Rompaey K,DHaens J. Prognostic value of perfusion-weighted imagingin brain glioma: a prospective study. Acta Neurochir (Wien).2006;148:277-85.Lupo JM, Cha S, Chang SM, Nelson SJ. Dynamic suscep-tibility-weighted perfusion imaging of high-grade gliomas:characterization of spatial heterogeneity. Am J Neuroradiol .2005;26:1446-54.

Calli C, Kitis O, Yunten N, Yurtseven T, Islekel S, Akalin T.Perfusion and diffusion MR imaging in enhancing malignantcerebral tumors. Eur J Radiol . 2006;58:394-403.Basser PJ, Jones DK. Diffusion-tensor MRI: theory, experi-mental design and data analysis a technical review. NMRBiomed . 2002;15:456-67.Goebell E, Fiehler J, Ding XQ, et al. Disarrangement offiber tracts and decline of neuronal density correlate in gliomapatientsa combined diffusion tensor imaging and 1H-MRspectroscopy study. Am J Neuroradiol . 2006;27:1426-31. Yu CS, Li KC, Xuan Y, Ji XM, Qin W. Diffusion tensor trac-tography in patients with cerebral tumors: a helpful techniquefor neurosurgical planning and postoperative assessment. Eur J

Radiol . 2005;56:197-204.Provenzale JM, McGraw P, Mhatre P, Guo AC, Delong D.Peritumoral brain regions in gliomas and meningiomas: inves-tigation with isotropic diffusion-weighted MR imaging anddiffusion-tensor MR imaging.Radiology . 2004;232:451-60.Li X, Vigneron DB, Cha S, et al. Relationship of MR-derivedlactate, mobile lipids, and relative blood volume for gliomas in vivo. Am J Neuroradiol . 2005;26:760-9.Hall WA, Martin A, Liu H, Truwit CL. Improving diag-nostic yield in brain biopsy: coupling spectroscopic targeting with real-time needle placement. J Magn Reson Imaging .2001;13:12-5.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Gajewicz W, Grzelak P, Grska-Chrzastek M, ZawirskKumierek J, Stefaczyk L. The usefulness of fused Mand SPECT images for the voxel positioning in proton netic resonance spectroscopy and planning the biopsy tumors: presentation of the method. Neurol Neurochir Pol .2006;40:284-90.

Graves EE, Nelson SJ, Vigneron DB, et al. Serial protospectroscopic imaging of recurrent malignant gliomas gamma knife radiosurgery. Am J Neuroradiol . 2001;22:613-24Sanai N, Berger MS. Glioma extent of resection and itson patient outcome. Neurosurgery . 2008;62:753-64.Nabavi A, Black PM, Gering DT, et al. Serial intraopertive magnetic resonance imaging of brain shift. Neurosurgery .2001;48:787-97.Nimsky C, Ganslandt O, Cerny S, Hastreiter P, GreinerFahlbusch R. Quantification of, visualization of, and csation for brain shift using intraoperative magnetic resoimaging. Neurosurgery . 2000;47:1070-9.Claus EB, Horlacher A, Hsu L, et al. Survival rates in with low-grade glioma after intraoperative magnetic reimage guidance.Cancer . 2005;103:1227-33.Kekhia H, Colen RR, Oguro S, et al. Investigating throle of intraoperative MRI in glioma resect ion: a volrical analysis. Presented at: 2010 Annual Meeting ofCongress of Neurological Surgeons; October 16-21, San Francisco, Calif. Walsh AR, Schmidt RH, Marsh HT. Cortical and localthetic resection as an aid to surgery of low and intermegrade gliomas.Br J Neurosurg . 1992;6:119-24.Berger MS, Kincaid J, Ojemann GA, Lettich E. Brain ping techniques to maximize resection, safety, and sei-

zure control in children with brain tumors. Neurosurgery .1989;25:786-92.Black PM, Ronner SF. Cortical mapping for defining thlimits of tumor resection. Neurosurgery . 1987;20:914-9.Meyer FB, Bates LM, Goerss SJ, et al. Awake craniotoaggressive resection of primary gliomas located in eloqbrain. Mayo Clin Proc . 2001;76:677-87.Nabavi A, Goebel S, Doerner L, Warneke N, Ulmer S,Mehdorn M. Awake craniotomy and intraoperative maresonance imaging: patient selection, preparation, and nique.Top Magn Reson Imaging . 2009;19:191-6.Pamir MN, zduman K. 3-T ultrahigh-field intraoperaMRI for low-grade glioma resection. Expert Rev Anticancer

Ther . 2009;9:1537-9.Hatiboglu MA, Weinberg JS, Suki D, et al. Impact of inoperative high-field magnetic resonance imaging guidaglioma surgery: a prospective volumetric analysis. Neurosurger.2009;64:1073-81.Pamir MN, zduman K, Dincer A, Yildiz E, Peker S, OMM. First intraoperative, shared-resource, ultrahigh-fi Tesla magnetic resonance imaging system and its applilow-grade glioma resection. J Neurosurg . 2010;112:57-69.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.


11/13



MAGNETIC RESONANCE IMAGING OFBRAIN TUMOR POST TESTExpires: July 15, 2013 Approved for 1 ARRT Category A Credit.

1. What are brain tumors?

Areas of reduced blood flow Abnormal proliferations of cellsSwollen tissuesBuildup of cerebrospinal fluid

2. Unlike secondary brain tumors, primary braintumors

can spread to other body parts and are alwaysmalignant.can spread to other body parts and can bemalignant or benign.remain confined to the brain and are alwaysmalignant.remain confined to the brain and can be malig-nant or benign.

3. Which of the following is NOT true of gliomas?Patient survival is very low.Relapse rates are high.Gliomas can easily be distinguished fromhealthy tissue.Gliomas are the most common primary malig-nant brain tumors.

4. MRI is the imaging study of choice for brain tumor because it

provides detailed information about lesion type,size and location.

is more widely available than other imagingmodalities.is inexpensive and time saving.is less painful than other imaging modalities.

5. Which of the following DOES NOT contributeto MRI signal generation?

T2-relaxation timeVoxelProton-density T1-relaxation time

6. An MRI signal is produced when transverse mag-netization is

zero.

switched off.different from zero.applied along the y-axis.

7. Gliomas appear in gadolinium-enhanced T1- weighted MR images as _______, _______ areas.

hyperintense; dark hypointense; dark hyperintense; brighthypointense; bright

8. Gliomas appear in T2-weighted MR images as_________, __________ areas.

hyperintense; dark hypointense; dark

a. b.c.d.

a.

b.

c.

d.

a. b.c.

d.

a.

b.

c.d.

a. b.c.d.

a.

b.c.d.

a. b.c.d.

a. b.

hyperintense; brighthypointense; bright

9. Vasogenic edema isa chronic subdural fluid collection.the accumulation of extracellular cerebrosfluid due to disruption of the blood-brain bthe benign extracerebral fluid collections are common in infants. when the BBB remains intact; it is due toderangement in cellular metabolism.

10. MRS is primarily usedto compensate for brain shift during intraotive studies.to determine the differential diagnosis of gfor tissue metabolism studies and for diffeating between tumor types.for the imaging of white matter.

11. In intraoperative MRI (iMRI), images are takenimmediately before surgery.immediately after surgery.immediately before, during, and after surgimmediately before and after surgery.

12. Compared to other imaging modalities, iMRI ismore effective at maximizing surgical resection while reducing the risk of damage to healthy tissue because it

does not require preoperative imaging stucan compensate for brain shift.can be used alone.allows for the use of ferromagnetic tools.

13. By maximizing resection, iMRI improves survival.

In one study, patients with partial glioma resection were _______ more likely to die within three yearsthan patients with total glioma resection.

4.9 times2.6 times1.8 times0.5 times

14. In another study, use of iMRI increased totalresection of non-infiltrating gliomas from ____to____ percent.

60; 10060; 8560; 80

60; 7515. Besides improving total surgical resection rates, when combined with cortical stimulation, bothMRI and iMRI can reduce the risk of

side effects from anesthetics.damage to eloquent areas.artifacts.patients losing consciousness.

16. To prevent iMRI from altering the results ofcortical stimulation during awake craniotomy, theoperation must be performed

with low-field MRI systems. with high-field MRI systems.

c.d.

a. b.

c.

d.

a.

b.c.

d.

a. b.c.d.

a. b.c.d.

a. b.c.d.

a. b.c.

d.

a. b.c.d.

a. b.


12/13



inside the magnetic field generated by the scanner.outside the magnetic field generated by the scanner.

17. iMRI is possible withlow-field (0.5 Tesla) and high-field (1.5 Tesla)systems.high-field (1.5 Tesla) and ultrahigh-field(3 Tesla) systems.low-field (0.5 Tesla), high-field (1.5 Tesla), andultrahigh-field (3 Tesla) systems.low-field (0.5 Tesla) and ultrahigh-field(3 Tesla) systems.

18. Compared to low-field system iMRI, high-fieldiMRI is associated with

improved image quality. worse signal-to-noise ratio.reduced speed.lower anatomical resolution.

19. One study found that high-field (1.5-T) iMRIincreased the extent of glioma resection from______ to _____ percent

76; 9676; 9074; 9074; 88

20. In another study, ultrahigh-field (3 Tesla) iMRIincreased the number of total glioma resections by______ percent.

10.318.632.344.8

c.d.

a.

b.

c.

d.

a. b.c.d.

a. b.c.d.

a. b.c.d.


13/13

Enterprises for Continuing Education, Inc.PO Box 300Brighton, MI 48116-0300Phone: 810-229-3354 Fax: 810-229-3235E-mail: [email protected] Website: www.cewebsource.com

CEwebsource.com ANSWER SHEETApproved by the AHRA for 1 Category A CE Credit

Please Note: Approved for ARRT and NMTCB Direct Credit

Circle the letter corresponding to the correct answer for each question.Mail or fax this completed answer sheet along with payment to the address or fax number on the top of thcredit card information must be included. You must receive a score of 75% or better to receive credit in any section. Allow up to 4 weeks to procesContinuing Education will be sent to you.Include payment. Answer keys must be accompanied by a $10 processing fee.In a hurry? RUSH SERVICE is available for an additional $10 for CREDIT CARD ORDERS. Fax this an with your credit card information to (810) 229-3235 for a 48-hour (M-F) turn-around! Whether faxing onseveral, only one $10 charge is added to the total of your order when faxing multiple sheets at once!

YES , in addition to the standard processing fee of $10, please charge my credit card account $10 EXTRA for RUSHSERVICE . FAX my expedited record of Continuing Education to me at: (______) _____-_______.

(If a fax number is not provided, a copy will be sent to the address indicated below within 48 hours)

Method of Payment: (checks payable to ECEI ) Check q Money Order q Visa q MasterCard q Discover q AmEx

Account Number Expires _____ / _____

dentification Section (Please print legibly in blue or black ink; do not leave blank):Name _____ __ __ __ __ __ __ __ __ __ __ __ __ ____ __ __ __ __ __ _ Email ___ __ __ __ ____ __ __ __ __ __ __ __ __ __ __

Address ______ ______ ______ _______ ______ ______ ______ _______ ______ _ Birth Month _______ ______ ____

City _____________ _______ State _____ Zip _______ _ Daytime Phone: ( __ __ __ ) __ __ __ __ __ __ __

Please check ONE:revious Name/Zip if changed since your last submission q MAIL my Record of Continuing Educationq E-MAIL my Record of Continuing Education

V-Code ______ ____ __

Article Title: Magnetic Resonance Imaging of Brain Tumor

1. a b c d 6. a b c d 11. a b c d 16. a b c d





CEwebsource.com ANSWER KEY MAGNETIC RESONANCE IMAGING OF

BRAIN TUMOR EXPIRES JULY 15, 2013

Office Use: MR71/

Documents

1809558_vkRJSgG6YNXYheolDuv5em1pV