Vascular Disorders—Magnetic Resonance Angiography: Brain Vessels




Magnetic resonance angiography (MRA) of the brain obtained at 3 T imaging has made a significant clinical impact. MRA benefits from acquisition at higher magnetic field strength because of higher available signal-to-noise ratio and improved relative background suppression due to magnetic field strength–related T1 lengthening. Parallel imaging techniques are ideally suited for high-field MRA. Many of the developments that have made 3 T MRA of the brain successful can be regarded as enabling technologies that are essential for further development of 7 T MRA, which brings additional challenges.








  • MRA obtained at 3 T is uniformly better than MRA obtained at 1.5 T, with major benefits including nearly double the SNR as well as magnetic field strength–related T1 lengthening, which improves relative background suppression.



  • Parallel imaging techniques are ideally suited for high-field MRA, with advantages including faster acquisition, reduced motion artifacts, decreased blurring, less geometric distortion related to susceptibility effects, and increased spatial resolution.



  • CE-MRA, and especially time-resolved CE-MRA, benefits substantially from acquisition at high magnetic field strength, with inherent competing requirements of both high spatial and temporal resolution.



  • Acquisition of both unenhanced 3D-TOF and CE-MRA sequences during the same imaging session is frequently advantageous, with the diagnostic potential of the combination of techniques better than either alone, especially for evaluation of pathologic vascular entities such as AVMs, DAVFs, and coiled aneurysms.



Key Points






  • Ischemic stroke and arterial occlusive disease



  • Aneurysm




    • Aneurysm screening/follow-up of untreated aneurysm



    • Aneurysm in the setting of acute subarachnoid hemorrhage



    • Treated (coiled) aneurysm



    • Giant/partially thrombosed aneurysm




  • Intracranial arterial dissection



  • Vascular malformations of the brain




    • Arteriovenous malformation



    • Capillary telangiectasia



    • Cavernous angioma



    • Developmental venous anomaly




  • Dural arteriovenous fistula



  • Vasculitis and vasospasm



  • Moyamoya



  • External carotid circulation pathology



Diagnostic Checklist
Over the past decade, there have been substantial software and hardware improvements that improved the quality of magnetic resonance angiography (MRA), with the introduction of routine clinical 3 T imaging likely the most significant. Additional improvements include parameter optimization, use of innovative k-space sampling schemes, development of improved imaging coil arrays, incorporation of parallel imaging methods, new gadolinium agents (eg, blood pool agents, higher relaxivity agents), and improvements in radiofrequency (RF) transmission (lower power pulses, parallel RF excitation). Many of these developments can be regarded as enabling technologies for high magnetic field MRA at 3 T and are essential for further development of 7 T MRA, which brings additional challenges. Parameters for commonly used MRA techniques are shown in Tables 1–4 .

Table 1

Comparison of 3D-TOF brain MRA techniques. Over the past decade, acquisition times for routine 3 T 3D-TOF MRA of the brain have decreased substantially, going from approximately 12 minutes to just more than 3 minutes, with comparable image quality and coverage
































































MR Imaging Scanner Acquisition Time (min) Matrix Partitions, Thickness, Overlap Repetition Time/Echo Time (millisecond) Flip Angle FOV (cm) Bandwidth (kHz) Imaging Options
1.5 T 11:28 256 × 224
1 NEX
145
1.4 mm
−0.7 mm
36/6.9 Fr 25° 18 × 16.2 15.6 VB, MT, Z512, Z2
3-slab MOTSA
3 T, 2001 protocol, unaccelerated 12:06 384 × 224
1 NEX
145
1.4 mm
−0.7 mm
38/4 Fr 25° 18 × 16.2 15.6 VB, MT, Z512, Z2
3-slab MOTSA
3 T, unaccelerated 6:08 384 × 224
1 NEX
145
1.4 mm
−0.7 mm
38/4.2 Fr 25° 18 × 16.2 15.6 ED, MT, Z512, Z2
3-slab MOTSA
3 T, accelerated, sense = 2, 8 channel 3:15 384 × 224
1 NEX
145
1.4 mm
−0.7 mm
25/3.9 Fr 25° 18 × 16.2 15.6 ED, MT, Z512, Z2
3-slab MOTSA
3 T, accelerated, GRAPPA = 2, 12 channel 4:24 384 × 235
2 NEX
128
0.7 mm
0 mm
21/3.69 25° 18 × 16 40.5 PFP
SAT1
5-slab MOTSA

Abbreviations: ED, extended dynamic range; FOV, field of view; Fr, fractional echo; GRAPPA, generalized autocalibrating partially parallel acquisition; MOTSA, multiple overlapping thin slab acquisition ; MT, magnetization transfer; NEX, number of excitation; PFP, phase partial fourier; SAT1, shifting parallel presaturation, applied to single side of acquisition slice; VB, variable bandwidth.


Table 2

Parameters for first-pass intracranial CE-MRA using bolus contrast administration and elliptic centric phase encoding order to capture contrast from opacified arteries at center of k-space. Timing can be done using a small timing bolus or automated triggering software


































MR Imaging Scanner Acquisition Time (min) Matrix Partitions, Thickness, Overlap Repetition Time/Echo Time (millisecond) Flip Angle FOV (cm) Bandwidth (kHz) Imaging Options
1.5 T 00:53 256 × 224
1 NEX
89
1.2 mm
−0.6 mm
6.6/2.4 45° Axial
22 × 22
31.2 ZIP 512
3 T 00:53 416 × 224
1 NEX
89
1.2 mm
−0.6 mm
6.7/2.06 40° Axial
22 × 22
31.2 ZIP 512

Abbreviations: NEX, number of excitations; ZIP, zero interpolate.


Table 3

CE-MRA parameters for steady state acquisition. This sequence may be acquired concomitantly with bolus contrast administration (eg, 0.1 mmol/kg standard gadolinium agent administered intravenously at 2–3 mL/s, followed by 15 to 20 mL saline at 2 mL/s). This CE-MRA technique can also be performed successfully if acquired promptly after prior intravenous gadolinium administration (eg, immediately after time-resolved CE-MRA)
























MR Imaging Scanner Acquisition Time (min) Matrix Partitions, Thickness Repetition Time/Echo Time (millisecond) Flip Angle FOV (cm) Bandwidth (kHz) Imaging Options
3 T 02:40 320 × 320
1 NEX
130
1.4 mm
5.5/1.4 30° Sagittal
25 × 25
62.5 ZIP 512, ZIP × 2, VB, ED

Abbreviations: ED, extended dynamic range; NEX, number of excitations; VB, variable bandwidth; ZIP, zero interpolate.


Table 4

Time-resolved CE-MRA parameters for commercially available TWIST and TRICKS techniques at 3 T. No timing bolus is required. For an adult with normal renal function, approximately 20 mL Multihance is administered intravenously at a rate of 3 mL/s, followed by a 20-mL saline flush given at 3 mL/s




























Time-Resolved MRA Technique Acquisition Time (min) Matrix
Orientation
Partition/Thick
Field of View
Repetition Time/Echo Time (millisecond) Flip Angle Frame Update (s) Other
TWIST
3 T
1:15 (0:07 for mask) 192 × 192
Sagittal
120/1.4 mm
26 × 26 cm FOV
3.15/1.27 25° 2.2 (30 × 3D volumes) 32 channel coil
GRAPPA 6× (Acceleration in 2 directions)
TRICKS
3 T
1:19 (0:09 for mask) 192 × 192
Sagittal
110/2.6 mm
0.75NEX
28 × 22 cm FOV
2.2/0.9 20 2.10 (36 × 3D volumes) 8 channel coil
Asset 2× (Acceleration in single direction)

Abbreviations: GRAPPA, generalized autocalibrating partially parallel acquisition; TRICKS, time-resolved imaging of contrast kinetics; TWIST, time-resolved angiography with interleaved stochastic trajectories.


Principle clinical indications for intracranial MRA include evaluation of ischemic stroke, arterial occlusive disease, aneurysms, cerebral vascular malformations, dural arteriovenous fistula (DAVF), central nervous system vasculitis, vasospasm, and moyamoya.


MRA evaluation of ischemic stroke and arterial occlusive disease


Causes of ischemic stroke include vascular thrombosis, cerebral embolism, hypotension, and anoxia/hypoxia. Three-dimensional time-of-flight (3D-TOF) MRA is the technique most commonly used for routine angiographic evaluation of the intracranial circulation in patients with ischemic stroke (cerebral infarction). Three-dimensional TOF MRA provides excellent depiction of the circle of Willis and its proximal large branches. It readily depicts thrombotic vascular occlusion as well as stenosis related to arterial occlusive disease manifested by processes such as atherosclerosis and dissection. Three-dimensional TOF MRA is quite accurate for depicting widely patent normal vessels or demonstrating complete vascular occlusion, although very-slow-flow or susceptibility artifact can mimic an occlusion. Accurate assessment of partial stenosis is more precise at 3 T than 1.5 T, especially for small vessels. Additional signal-to-noise ratio (SNR) available from acquisition at higher magnetic field strength in conjunction with improved multicoil arrays allows for 3D-TOF MRA with higher spatial resolution, parallel acquisition techniques with inherent reduction in acquisition time, as well as motion artifact ( Fig. 1 ). Parallel imaging techniques gives rise to potentially confounding reconstruction artifacts ( Fig. 2 ). In-plane saturation effects that can result in a tapered appearance of vessels because of signal dropout due to slower-moving peripheral blood are less prominent at 3 T. The combination of conventional and low-dose postgadolinium 3D-TOF MRA has been advocated to provide more robust and specific evaluation of intracranial vascular stenosis.




Fig. 1


Comparison of early 3 T 3D-TOF MRA with comparable 1.5 T 3D-TOF MRA, obtained using standard quadrature head coils. A decade ago, 1.5 T 3D-TOF MRA ( A ) and 3 T 3D-TOF MRA ( B ) typically required 10- to 12-minute acquisition times. Parallel acquisition techniques have been one of the most significant advances for conventional 3D-TOF MRA over the past decade, substantially reducing acquisition times. Axial maximum intensity projection (MIP) collapse from 3 T MRA using ASSET (array spatial sensitivity encoding technique) 2× acceleration acquired in 3:15 minutes using an 8-channel head coil array ( C ). Axial MIP collapse from 3 T MRA using GRAPPA (generalized autocalibrating partially parallel acquisition) 2× acceleration, acquired in 4:24 minutes using a 12-channel head coil array ( D ). An inherent benefit of acceleration techniques is reduction of patient motion artifact, principally due to decreased acquisition time. Intracranial MRA at 7 T remains primarily a research tool. Comparison of 7 T and 1.5 T 3D-TOF MRA of the anterior circle of Willis ( E , F ). Note excellent depiction of lenticulostriate arteries at 7 T, not visible on the 1.5 T MRA.

( Courtesy of Dr Zang-Hee Cho, Neuroscience Research Institute, Gachon University of Medicine and Science.)



Fig. 2


Three-dimensional TOF MRA parallel acquisition artifacts. Use of parallel imaging techniques can lead to unique image artifacts, which if unrecognized could potentially lead to diagnostic misinterpretation. MRA source images should always be reviewed. ( A , B ) Axial 3D-TOF MRA source images depicting typical “wrap” artifacts related to ASSET (array spatial sensitivity encoding technique) parallel imaging. Patient motion is a frequent cause for these artifacts.


Phase-contrast MRA (PC-MRA) techniques are not frequently used for evaluation of stroke, except if directional flow information or velocities are desired. The phase images of PC-MRA can be used to determine direction of collateral flow about the circle of Willis. PC-MRA is useful for performing MRA after administration of intravenous gadolinium because PC-MRA benefits from acquisition with presence of intravascular gadolinium, whereas 3D-TOF MRA is typically suboptimal following full-dose gadolinium administration because of confounding overlapping vascularity due to opacified venous structures.


For specific clinical indications, gadolinium contrast–enhanced MRA (CE-MRA), including time-resolved CE-MRA, can be helpful to assess intracranial vascular patency ( Fig. 3 ). High-resolution 3 T black blood MRA permits high-resolution vessel wall imaging useful for assessment of intracranial atherosclerosis.




Fig. 3


Three-dimensional TOF MRA and CAPR (Cartesian projection reconstruction) time-resolved CE-MRA were used to evaluate a 3-day-old infant with seizures, coagulopathy, and suspected hypoxic ischemic encephalopathy. ( A ) Conventional 3 T 3D-TOF MRA was obtained but nondiagnostic because of severe degradation by slab interface artifact as well as RF artifact arising from medication infusion pumps. The apparent intracranial stenoses noted on the 3D-TOF MRA within the left posterior cerebral artery ( solid arrow ) and left anterior and middle cerebral arteries on the 3D-TOF sequence ( open arrow ) were confirmed to represent artifact with a hand injected 1-mL CAPR time-resolved CE-MRA ( B ) that demonstrated normal patency of the intracranial vessels. ( C ) Cephalhematoma and overriding calvarial sutures ( long arrow ) identified on sagittal fast-spin echo T2-weighted image in the same patient, with clear depiction of flow defect within the superior sagittal sinus at the level of suture overlap ( long arrow ) ( D ). Sagittal venous phase MIP images from same CE-MRA shows antegrade filling of anterior superior sagittal sinus ( E , F ) ( double arrows ).




MRA evaluation of aneurysm


A common application of intracranial MRA is evaluation of arterial aneurysms, which can be categorized morphologically as either fusiform (10%) or saccular (90%). Saccular aneurysms most commonly arise from the circle of Willis (90%) and from the vertebrobasilar system (10%) and are believed to arise from flow-related vessel wall stresses and a complex combination of genetically inherited susceptibility. The incidence of sporadic intracranial aneurysms is 1% to 2% in autopsy series. There is an approximate 10% prevalence for familial intracranial aneurysms, which typically present in younger patients compared with those with sporadic aneurysms. Conditions predisposing aneurysm formation include autosomal dominant polycystic kidney disease, aortic coarctation, fibromuscular dysplasia, aberrant vascular anatomy (persistent trigeminal artery, arterial fenestration, azygous anterior cerebral artery), or connective tissue disorder (eg, Ehlers-Danlos). Aneurysms often develop on arterial feeders of an AVM; the causes of these are believed to be flow related. An additional 1% to 3% of aneurysms are related to trauma ( Fig. 4 ) and mycotic or oncotic etiology.




Fig. 4


Three-Tesla 3D-TOF MRA is useful for screening of aneurysms in asymptomatic individuals. Axial MIP collapse ( A ) and source image ( B ) depicting a posterior cerebral artery (PCA) aneurysm arising from the P2 segment ( arrows ). The location of this PCA aneurysm adjacent to the tentorial incisura is typical for a posttraumatic aneurysm, such as in this patient with a known history of prior severe motor vehicle accident.


Specific subindications for MRA assessment of aneurysms include screening for aneurysms in asymptomatic high-risk populations, detection of aneurysms in the setting of acute subarachnoid hemorrhage (SAH), follow-up of known untreated aneurysms, follow-up of treated endovascularly coiled aneurysms, and evaluation of giant or partially thrombosed aneurysms. The optimal MRA technique used to evaluate aneurysms in these differing settings varies with the specific indication, as discussed later.







  • Ischemic stroke and arterial occlusive disease



  • 3D-TOF MRA




  • Aneurysm



  • Aneurysm screening, follow-up of untreated aneurysm: 3D-TOF MRA



  • Aneurysm in setting of acute subarachnoid hemorrhage: CTA, DSA, 3D-TOF MRA (typically not first study, if obtained)



  • Treated (coiled) aneurysm: 3D-TOF MRA and intracranial CE-MRA



  • Giant/partially thrombosed aneurysm: 3D-TOF MRA, PC-MRA, CE-MRA




  • Intracranial arterial dissection



  • 3D-TOF MRA, pregadolinium fat-suppressed T1 spin-echo, black blood MRA




  • Vascular malformations of the brain



  • Arteriovenous malformation: 3D-TOF MRA, a PC-MRA, CE-MRA, time-resolved CE-MRA



  • Capillary telangiectasia: GRE/SWI, postgadolinium T1 imaging b



  • Cavernous angioma: GRE/SWI, postgadolinium T1 imaging b



  • Developmental venous anomaly: CE-MRV, postgadolinium T1 imaging b




  • Dural arteriovenous fistula



  • 3D-TOF MRA, a CE-MRA, time-resolved CE-MRA




  • Central nervous system vasculitis



  • 3D-TOF MRA, conventional MR imaging (GRE/SWI, T2/FLAIR, pregadolinium + postgadolinium T1 imaging)




  • Vasospasm



  • 3D-TOF MRA, CE-MRA, perfusion imaging




  • Moyamoya



  • 3D-TOF MRA a , CE-MRA, time-resolved CE-MRA




  • External carotid circulation pathology



  • CE-MRA, time-resolved CE-MRA



Abbreviations: 3D-TOF, three-dimensional time of flight; CE-MRA, gadolinium contrast–enhanced magnetic resonance angiography; CTA, computed tomographic angiography; DSA, digital subtraction angiography; FLAIR, fluid attenuated inversion recovery; GRE, gradient echo; MRA, magnetic resonance angiography; MRV, magnetic resonance venography; PC-MRA, phase-contrast magnetic resonance angiography; SWI, susceptibility weighted imaging.


a Acquisition of both unenhanced 3D-TOF and CE-MRA sequences during the same imaging session is frequently advantageous, with the diagnostic potential of the combination of techniques better than either alone, especially for evaluation of pathologic vascular entities such as arteriovenous malformations, dural arteriovenous fistulas, and coiled aneurysms.


b Conventional postgadolinium T1-weighted imaging (eg, SE, GRE, MP-RAGE, and so forth).


Recommended pulse sequence chart: 3 T magnetic resonance angiography of the brain


MRA Screening and MRA Observation of Untreated Aneurysms


In the nonacute setting, 3D-TOF MRA is an excellent tool for following known aneurysms and screening for aneurysms in high-risk populations, such as for individuals with known autosomal dominant polycystic kidney disease or aortic coarctation. Superior depiction of aneurysms using 3D-TOF MRA at 3 T versus 1.5 T has been demonstrated. Using 3D-TOF MRA, aneurysms as small as 2 mm can be detected with a sensitivity of 74% to 98%. More recent literature has reported the possibility of detection of aneurysms as small as 1 mm ; however, to date, no large study examining the sensitivity and specificity for detection of aneurysms that small has been performed. Aneurysm size and location influence MRA sensitivity for detection of unruptured untreated aneurysms. Recently developed inversion recovery–based MRA techniques ( Fig. 5 ) and CE-MRA ( Fig. 6 ) are used less commonly than TOF techniques but are sensitive for detection of aneurysms.




Fig. 5


Fast inversion recovery (FIR) MRA techniques combine use of inversion recovery methods and acceleration techniques made possible by 3 T MR imaging. This example demonstrates an asymptomatic right middle cerebral artery trifurcation aneurysm with unenhanced 3D-TOF MRA ( A, arrow ) and FIR-MRA ( B, arrow ). This FIR-MRA acquisition used a self-calibrated parallel imaging technique, with an acceleration factor of 2. ( C ) A second aneurysm arises from the medial aspect of the left cavernous internal carotid artery (ICA), which was not prospectively identified on CTA ( arrow , top ). This cavernous ICA aneurysm is well identified with the TOF technique ( arrow , middle ), but the cavernous region can have venous and fat signal that can confound diagnosis. Signal saturation leads to slight signal loss within the aneurysm lumen with the TOF MRA. With the FIR technique, however, the cancellation of background signal makes both aneurysms stand out ( arrow , bottom ).



Fig. 6


Aneurysm detection with CAPR (Cartesian projection reconstruction) time-resolved CE-MRA. Although the CAPR MRA technique was primarily developed for evaluation of dynamic vascular pathology, including vascular malformations or partial stenosis, it can also detect aneurysms. CAPR CE-MRA (4× acceleration, image resolution of 0.9 mm × 1.3 mm × 2 mm, frame update rate of 2.3 seconds) depicting a 3-mm aneurysm projecting laterally from the right paraclinoid internal carotid artery ( arrow , A ). Corresponding DSA ( arrow , B ) and volume-rendered CTA images ( arrow , C ) confirming presence of the aneurysm.


Digital subtraction angiography (DSA) is generally a safe procedure but still has a small but nonzero risk of morbidity, including death. A prior large review has shown an overall neurologic morbidity rate of 1.3% and a permanent neurologic complication rate of 0.5%. Risk factors for DSA included advancing age (>55 years), preexisting cardiovascular disease, and longer duration of DSA procedure. For patients allergic to iodinated contrast material, DSA may be performed after pretreatment with steroids, such as in the Lasser protocol.


Aneurysms in the Setting of Acute SAH


The incidence of SAH is approximately 6 to 8 per 100,000 persons per year, with peak incidence occurring in the sixth decade of life. Most nontraumatic SAHs are the result of an intracranial aneurysm rupture.


MRA is typically not the first study in a patient with acute SAH. In the acute setting of SAH, conventional DSA remains the generally accepted gold standard for detection of intracranial aneurysms. Computed tomographic angiography (CTA) is currently playing an increasingly prominent role in the evaluation of acute SAH because it is convenient to perform CTA immediately after detection of an SAH on noncontrast head computed tomography (CT). CTA is usually immediately available, whereas magnetic resonance (MR) imaging often requires patient transfer to the MR imaging suite, which may not be near the emergency department. In addition, it is frequently necessary to summon on-call personnel necessary for MR imaging operation after routine hours. CTA has also proved quite accurate, with a reported success rate of 98% to 100% in detecting aneurysms in the setting of SAH.


There remains a role for MRA in the setting of negative DSA examination results because false-negative rates of up to 5% to 10% are reported with DSA. It is known that vasospasm, thrombosis of the aneurysm or of the parent vessel, compression of the aneurysm by adjacent blood or edema, small aneurysm size, and limited projections may obscure a ruptured aneurysm on DSA. In these situations, MRA has been shown to find aneurysms not identified on prior DSA.


MRA Evaluation of Treated (Coiled) Aneurysms


MRA is emerging as the technique of choice for long-term evaluation of residual or recurrent patency of aneurysms treated with endovascular coiling. Platinum alloy coils are designed to allow optimum visibility during endovascular treatments performed under fluoroscopic control. However, although this high attenuation is desirable during endovascular placement of the coils, the endovascular coil mass results in substantial beam hardening and streak artifact on subsequent CT and CTA examinations, degrading depiction of the aneurysm, adjacent vessels, and surrounding brain parenchyma. The susceptibility artifact arising from these platinum alloys is less at 3 T than 1.5 T predominately because of the ability to use smaller voxel sizes, usually resulting in only mild distortion of the local magnetic field and only slight loss of MRA signal intensity ( Fig. 7 ). Because of these factors, MRA is favored over CTA for follow-up of coiled aneurysms. Unlike platinum coils, metallic aneurysm clips usually demonstrate substantial magnetic susceptibility artifact that usually precludes adequate MRA assessment of residual/recurrent aneurysm for surgically treated aneurysms ( Fig. 8 ).




Fig. 7


Platinum alloy coils used for endovascular treatment of aneurysms are designed for optimal visibility during deployment under fluoroscopic control. The endovascular coil mass often results in substantial beam hardening and streak artifact on subsequent CT and CTA examinations ( A ), degrading depiction of the aneurysm, adjacent vessels, and surrounding brain parenchyma. ( B ) A 3 T CE-MRA MIP collapse and ( C ) a 3 T 3D-TOF MRA MIP collapse from same patient as in ( A ) depicting large aneurysm remnant ( block arrows ) of previously treated coiled aneurysm. Susceptibility artifact associated with platinum coils is relatively small, even at 3 T, usually resulting in only negligible distortion of the local magnetic field and only slight loss of MRA signal intensity.



Fig. 8


Unlike platinum coils, magnetic susceptibility artifacts related to a surgically clipped aneurysm are usually quite pronounced, especially at 3 T. Anteroposterior (AP) radiograph from a DSA study ( A ) showing changes of left craniotomy with prior surgical clipping of a left middle cerebral artery aneurysm ( block arrow , 3 clips) and endovascular coiling of a superior cerebellar artery aneurysm ( arrow ). Source image of 3D-TOF MRA ( B ) and targeted maximum intensity projection (MIP) image ( C ) from same patient showing susceptibility artifact “blowout” and corresponding loss of signal on the MIP image. Artifact related to aneurysm clips usually precludes adequate MRA assessment of residual/recurrent aneurysm. ( D ) Coiled left superior cerebellar artery aneurysm ( arrow ) has only minimal associated susceptibility artifact. Also note additional untreated aneurysm arising from the left cavernous internal carotid artery ( small arrows ). The tiny “dog-ear” aneurysm remnant ( arrowhead ) is seen on the DSA ( E ) and rotational DSA ( F ) and also on both CE TOF MRA ( G ) and CE-MRA ( H ) sequences.


Evaluation of Giant Aneurysms and Partially Thrombosed Aneurysms


An aneurysm exceeding 2.5 cm in diameter is considered a giant aneurysm. These typically occur in the cavernous internal carotid arteries and middle cerebral artery (MCA) bifurcations. Giant aneurysms have a higher rate of rupture than smaller saccular aneurysms. Giant aneurysms are believed to enlarge from recurrent internal hemorrhage and characteristically have laminated mural hemorrhages of varying ages.


MRA evaluation of giant aneurysms is often not ideal with 3D-TOF MRA because of saturation effects on slow or recirculation internal flow ( Fig. 9 ). They are typically easily identified on standard imaging and can be visualized with other MRA techniques such as CE-MRA or PC-MRA. Although less marked at 3 T, slow and/or turbulent flow within the lumen of an aneurysm may lead to dramatic signal loss on conventional 3D-TOF MRA with associated reduction in detection with this technique. Also, a thrombosed aneurysm may be harder to characterize with TOF MRA because the thrombus can be isointense. Alternatively, subacute thrombus within the wall or periphery of the aneurysm can also result in a confounding bright T1 “shine-through” signal, resulting in misinterpretation of overall patency of the aneurysm lumen. Complete clinical evaluation of a TOF MRA should include direct review of the source images, as exclusive review of the maximum intensity projection (MIP) images alone is often diagnostically insufficient. CTA is also very useful for evaluation of giant aneurysms. Research applications of high-field PC-MRA include computational simulation of flow dynamics in a giant intracranial aneurysm.




Fig. 9


MRA evaluation of a coiled supraclinoid internal carotid artery aneurysm. DSA image depicting an aneurysm before treatment ( A ) and later DSA image following endovascular coiling ( B ). A 3-month follow-up MRA was performed: 25 mm thick sagittal maximum intensity projections are shown on ( C ) 3D-TOF MRA, ( D ) CE-MRA, ( E ) and fast inversion recovery (FIR)-MRA. CE-MRA shows the upper and lower aneurysm remnants ( arrows ) better than TOF MRA. The appearance of the remnant on FIR-MRA is nearly as good as on the CE-MRA. Also, note that vessel conspicuity in FIR-MRA is superior to that in TOF MRA in this 76-year-old patient. Note that the platinum coils do not result in substantial susceptibility artifact on the MRA acquisitions, even though acquired at 3 T.




MRA evaluation of aneurysm


A common application of intracranial MRA is evaluation of arterial aneurysms, which can be categorized morphologically as either fusiform (10%) or saccular (90%). Saccular aneurysms most commonly arise from the circle of Willis (90%) and from the vertebrobasilar system (10%) and are believed to arise from flow-related vessel wall stresses and a complex combination of genetically inherited susceptibility. The incidence of sporadic intracranial aneurysms is 1% to 2% in autopsy series. There is an approximate 10% prevalence for familial intracranial aneurysms, which typically present in younger patients compared with those with sporadic aneurysms. Conditions predisposing aneurysm formation include autosomal dominant polycystic kidney disease, aortic coarctation, fibromuscular dysplasia, aberrant vascular anatomy (persistent trigeminal artery, arterial fenestration, azygous anterior cerebral artery), or connective tissue disorder (eg, Ehlers-Danlos). Aneurysms often develop on arterial feeders of an AVM; the causes of these are believed to be flow related. An additional 1% to 3% of aneurysms are related to trauma ( Fig. 4 ) and mycotic or oncotic etiology.


Mar 20, 2017 | Posted by in NEUROLOGICAL IMAGING | Comments Off on Vascular Disorders—Magnetic Resonance Angiography: Brain Vessels

Full access? Get Clinical Tree

Get Clinical Tree app for offline access