Dr. Gotwald is a Fellow and Dr. Beauchamp is an Assistant Professor in the Neuroradiology Division, Russel H. Morgan Department of Radiology, Johns Hopkins Medical Institutions, Baltimore, MD.

Although catheter angiography remains the gold standard, non-invasive angiographic techniques are increasingly used for assessment of the neurovasculature. Computed tomographic angiography (CTA) and magnetic resonance angiography (MRA) are particularly appealing due to their low morbidity and comprehensiveness. The main goal of this article is to clarify the role of CTA and MRA in the evaluation of intracranial vascular disease by discussing the relative strengths and weaknesses of both techniques in the context of specific neurovascular disorders.

Imaging and post-processing methods

CT angiography--With modern spiral CT scanners it is possible to scan with continuing gantry rotation and patient table movement. High capacity x-ray tubes and faster (subsecond) tube rotations allow rapid acquisitions of very accurate volume data sets and enable scan completion during the period of optimal intravascular contrast enhancement.

In a standard CTA procedure the patient is scanned in the supine position and axial orientation. In general, 2-mm beam collimation and 3 mm/s table feed are sufficient, although smaller reconstruction intervals can be afforded with newer generations of CT scanners or multidetector equipment. The scanning parameters depend on the anatomic region being imaged. The scan volume should be restricted to the region of interest.

Optimization of vascular opacification is an essential prerequisite for CTA. An attenuation coefficient of approximately 150 HU is considered optimal for evaluation of vessel diameter. ¹ This can be obtained by using a power injector and an injection rate of 3 mL/sec. The time interval between the start of contrast infusion and initiation of the scan sequence can either be determined empirically or with automated techniques (e.g. SMART Prep [GE Medical Systems, Milwaukee, WI], CARE bolus [Siemens Medical Systems, Iselin, NJ]). However, all these methods seem to be equally reliable with the former approach somewhat more time efficient. ¹ With these settings, the entire spiral CT study can be accomplished in <1 minute. The image data can be reconstructed to obtain an angiographic display. The reconstruction algorithms and display methods will be discussed later. Fundamentally, CTA is an 'anatomic' study, similar to catheter angiography, in which a patent vessel will fill with contrast that will be subsequently incorporated into the vascular reconstruction.

MR angiography--Magnetic resonance angiography is a "physiologic" study in which generation of intravascular signal is also impacted by flow velocity, turbulence, and vascular orientation. The two fundamentally different methods are time-of-flight (TOF) and phase-contrast (PC) MRA. MRA enables vascular delineation with either bright (flow increases signal) or black (flow decreases signal) blood techniques. TOF MRA is used far more often. Multiple radiofrequency (RF) pulses are applied in a single acquisition that result in suppression of signal from stationary tissue. Areas not saturated, i.e., blood flowing into the plane of imaging, leads to the generation of intravascular signal differentiable from background tissue. Three-dimensional (3D) TOF acquisitions with thinner volume slabs afford greater signal to-noise ratios and spatial resolution than two-dimensional (2D) TOF. The quality of TOF MRA continues to improve, including optimization of background suppression using magnetization transfer (intracranial only), and minimization of turbulence- and saturation-related artifactual signal suppression using multiple overlapping thin-slab acquisition and tilted optimized non-saturated excitation. Finally, although TOF MRA is relatively insensitive to slow flow due to the RF saturation of the slow flow spins, paramagnetic contrast agents can be applied to increase the signal amplitude and, thus, improve vessel display.

The notable limitation of TOF MRA is that in addition to areas of flow, tissue with very short T1 will not be suppressed during repetitive stimulation (e.g., methemoglobin or fat) and can be falsely incorporated into the vascular reconstruction.

In these cases phase-contrast (PC) MRA should be applied. PC MRA uses phase shift differences induced by velocity of flow to create an angiographic image. Stationary tissue is subtracted from two acquisitions with bipolar velocity gradients. Thus, only tissue that has changed position between the two acquisitions, regardless of T1 relaxation properties, is included in the vascular reconstruction. This method affords excellent background suppression. However, two factors limit the more frequent application of PC MRA. The addition of the second acquisition (for subtraction) leads to increased evaluation time. Also, PC imaging requires predetermination of the range of expected velocities. Although easily done when flow is laminar, vascular tortuosity, numerous flow dividers, and the presence of stenosis limit the utility of PC for routine application.

Rendering and display techniques

Surface shaded display, maximum intensity projection, and volume rendering techniques are now feasible from a time and cost perspective due to continued advances in computer hardware, software, and display technology. They can be used for both CTA and MRA data. ²

Surface shaded display (SSD)--SSD was the first rendering technique applied to medical data sets. To segment the structures of interest for SSD a simple thresholding can be used. The user-specified range of attenuation values determines the intensity of each voxel within the data set. Advantages of SSD are superior speed as well as very accurate surface display. Multiple surfaces can be displayed when partial opacities are implemented. However, SSD only uses only approximately 10% of the available data. As a result, organ structures that do not have naturally differentiated surfaces are not well visualized.

Maximum intensity projection (MIP)­­MIP is widely used for angiographic display of MR and CT data. The highest voxel value from a projection through the dataset is displayed. This allows more accurate evaluation of the vessel diameter accuracy compared with SSD. The MIP technique is very useful for MRA and CTA display. ³ However, high-density material (i.e., calcifications) will obscure valuable information from intravascular contrast material. For this reason, MIP often requires extensive editing to eliminate unwanted data to create useful images.

Volume rendering (VR)--Neither SSD nor MIP displays vessel, parenchyma, and bone simultaneously. This only has been afforded by VR. This interactive technique allows for data processing and display in real time. ⁴ VR sums-up attenuation values of each voxel along parallel or divergent (i.e., perspective VR) lines through the data set for each pixel of the display. 3D VR has some advantages over the other rendering techniques. Most importantly, it incorporates the entire volume data set into a 3D image. Therefore, the dynamic range of the image is preserved, enabling simultaneous display of vessels, soft tissue, viscera, and bone. Window and center settings can be adjusted and clip planes can be applied to visualize structures in the volume that would otherwise be obscured. It is important to state that the superior editing capabilities of VR also present the risk for misrepresentation of vascular disease. Therefore, careful adjustment of the rendering parameters is critical. VR can be used to post-process MRA data sets. However, as noted above, with MRA the background voxel signals are suppressed for adequate vessel conspicuity. Bone and soft tissue surrounding the vessels is therefore poorly delineated. Thus the unique benefit of VR, on the fly evaluation of vessels and vascular relationships to skull base and brain tissue, is not taken advantage of with MRA.

Specific applications of CTA and MRA

Stroke--Stroke is the third most common cause of mortality in the United States. Efficacious therapy with thrombolytic agents is now available, but safe treatment hinges on the use of neuroimaging. Non-enhanced CT is valuable in excluding patients with extensive infarction or hemorrhage and has enabled treatment of patients presenting within 3 hours of stroke onset. In attempting to extend the window of treatment beyond 3 hours, emphasis has been placed on increasing the sensitivity and specificity in detecting vascular occlusion and areas of hypoperfusion. ^5-7 Vascular occlusion can be identified expeditiously with CTA or MRA with equal sensitivity. ^8-11 (figure 1) These techniques can also be supplemented with CT perfusion or MR perfusion imaging to delineate areas of hypoperfusion or tissue at risk for infarction. Although both CT- and MRI-based techniques are valuable, they have unique strengths. Specifically, CT is more readily available, less expensive, and less time consuming. The relatively wide gan-try opening and lack of magnetic field increases the ease of monitoring critically ill patients. Conversely, MRI is more sensitive in the detection of posterior fossa stroke and does not require the administration of intravenous contrast.

MR diffusion imaging enables early detection of areas of infarction minutes after the ictus. MR perfusion is also valuable to discriminate hypoperfused but still viable regions from non-salvable infarction zones. CT perfusion may afford similar delineation of reversible injury. CT perfusion is currently limited to one single slice. However, with multi-detector CT technology, whole brain coverage similar to MR perfusion should be attainable in the future.

The optimal approach in cases of stroke varies with the clinical situation and the availability of MRI technology. When MRI is immediately available, it is increasingly seen as the test of choice. If any delay in MRI evaluation is expected, a comprehensive CT evaluation with CT perfusion and CTA should be performed.

Dural sinus thrombosis--Although relatively uncommon, dural sinus thrombosis is a potentially life threatening condition. Predisposing factors are dehydration, hypercoaguable states, pregnancy, and intracranial infections. CTA and MRA are of value because the conventional CT and MRI findings (i.e., empty delta sign in CT) can be insensitive and nonspecific. MR venography is the technique of choice. ¹² It enables delineation of venous sinuses as well as sensitive detection of secondary parenchymal changes and other pathologic processes that may mimic sinus thrombosis symptoms.

CT venography is an alternative technique that enables excellent vascular delineation. In addition to being the study of choice when MR venography is not possible, it is also an excellent technique to assess interval change, response to treatment, and may be the only test required when clarifying a question of thrombus raised on a non-enhanced CT.

Intracranial aneurysms--Aneurysm rupture is associated with high morbidity and mortality. Increasing emphasis is being placed on the detection of aneurysms prior to rupture because elective aneurysm clipping and coiling can now be performed with extremely low patient risk. An ideal screening examination should have adequate sensitivity, minimal risk, and be "acceptable" to patients. ¹³ MRA and CTA are ideally suited for this purpose and both tests have equivalent sensitivity, i.e., detection of aneurysm >= 3 mm. ^14-23 The advantage of MRA in such cases is the ease of post-processing given the inherent background suppression. However, this is becoming less of an issue for CTA as volume rendering workstations equipped with clip planes enable physicians to interactively remove obscuring bone and adjacent vessels. Real-time 2D and 3D display with VR enables expedient evaluation for the presence of aneurysms. ²⁴

In addition to screening, CTA is uniquely valuable as a supplement to MRA and catheter angiography. For example, areas of turbulent flow, such as may be seen in giant aneurysms, result in loss of intravascular signal secondary to intra-voxel phase dispersion. CTA is "anatomic" and as such is relatively unaffected by complex flow patterns. (figure 2) Similarly with catheter angiography, overlapping vessels can obscure evaluation of aneurysm neck and dome morphology as well as its relationships to adjacent vessels. CTA enables a 3D delineation of vascular relationships to skull base and brain parenchyma not possible with either MRA or catheter angiography. This is an important issue in surgical planning and in determining the feasibility of endovascular coiling.

Increasingly, non-invasive techniques are used in the evaluation of patients with suspected aneurysms. For initial screening, MRA and CTA are essentially equivalent. In pretreatment planning, CTA is clearly superior due to its unique editing and display capabilities. Currently, it remains the standard of care to evaluate all aneurysm patients before surgery with catheter angiography so that aneurysms < 3 min are detectable. However, it can be expected that with continued technological advancements (such as the CT multi-detector array ²⁵ ), non-invasive techniques will offer sensitivity similar to catheter angiography.

Arteriovenous malformations (AVMs)-- Pial AVMs are congenital lesions with an estimated 2 to 4% risk of intraparenchymal hemorrhage. Hemorrhage is related to associated aneurysms, outflow restriction, or pure deep venous drainage. ^26-28 The goal of diagnostic evaluation is delineation of the nidus, the venous drainage patterns, associated aneurysms, and venous constrictions. Central to this differentiation is temporal resolution sufficient to determine whether a vessel is arterial or venous. CTA/venography is limited due to its low temporal resolution, as MRA/ venography is limited due to turbulent flow associated with these high-flow lesions. Thus, the role of non-invasive imaging is restricted
to certain instances. Specifically, CTA/ venography can be of valuable to define the morphology of the aneurysms, venous constrictions, and vascular relationships for surgical planning. (figure 3) Conventional MR imaging in combination with MRA/ venography is useful for assessing response to radiosurgery.

Conclusion

CTA and MRA techniques are highly valuable tools in many applications in the evaluation of vasculature in neuroradiology. It is hoped that this article will facilitate the use of non-invasive tests whenever appropriate. AR

References

1. Provenzale JM, Beauchamp NJ Jr.: Recent advances in imaging of cerebrovascular disease. Radiol Clin North Am 37:467-88, 1999.

2. Bullit E, Lie A, Aylward S, et al: Methods for displaying intracerebral vasculature anatomy. AJNR 18:417-420, 1997.

3. Napel S, Marks MP, Rubin GD, et al: CT angiography with spiral CT and maximum intensity projection. Radiology 185:607-610, 1992.

4. Kuszyk BS, Heath DG, Johnson PT, Johnson PT, et al: CT angiography with volume rendering for quantifying vascular stenosis: In vitro validation of accuracy. AJR AM J Roentgenol 173:449-455, 1999.

5. Beauchamp NJ Jr., Barker PB, Wang PY, van Zijl PC: Imaging of acute cerebral ischemia. Radiology 212:307-324, 1999.

6. Koenig M, Klotz E, Luka B, et al: Perfusion CT of the brain: Diagnostic approach for early detection of ischemic stroke. Radiology 209:85-93, 1998.

7. Sunshine JL, Tarr RW, Lanzieri CF, et al: Hyperacute stroke: Ultrafast MR imaging to triage patients prior to therapy. Radiology 212:325-332, 1999.

8. Fujita N, Hirabuki N, Fujii K, et al: MR imaging of middle cerebral artery stenosis and occlusion: Value of MR-angiography. AJNR 15:335-341, 1994.

9. Knauth M, von Kummer R, Jansen O, et al: Potential of CT-angiography in acute ischemic stroke. AJNR 18:1001-1010, 1997.

10. Shrier DA, Tanaka H, Numaguchi Y, et al: CT-angiography in the evaluation of acute stroke. AJNR 18:1011-1020, 1997.

11. Zawadzki MB: CT-angiography in acute ischemic stroke: The right tool for the job? AJNR 18:1021-1023, 1997.

12. Ouanounou S, Tornsick TA, Heitsman C, Holland CK: Cavernous sinus and inferior petrosal sinus flow signal on three-dimensional time-of-flight MR angiography. AJNR Am J Neuroradiol 20:1476-1481, 1999.

13. Raaymakers TW, Buys PC, Verbeeten B Jr., et al: MR angiography as a screening tool for intracranial aneurysms: Feasibility, test characteristics, and interobserver agreement. AJR Am J Roentgenol 173:1469-1475, 1999.

14. Alberico RA, Patel M, Caesy S, et al: Evaluation of the circle of Willis with three-dimensional CT-angiography in patients with suspected intracranial aneurysms. AJNR 16:1572-1578, 1995.

15. Aoki S, Sasaki Y, Machida T, et al: Cerebral aneurysms: Detection and delineation using 3D-CT-angiography. AJNR 13:115-1120, 1992.

16. Brown JH, Lustrin ES, Lev MH, et al: Characterization of intracranial aneurysm using CT angiography. AJR 169:889-893, 1997.

17. Hope JK, Wilson JL, Thomson FJ: Three-dimensional CT angiography in the detection and characterization of intracranial berry aneurysms. AJNR Am J Neuroradiol 17:439-445, 1996.

18. Ida M, Kurisu Y, Tamashita M: MR-angiography of ruptured aneurysms in acute subarachnoid hemorrhage. AJNR 18:1025-1032, 1997.

19. Kahara VJ, Seppanen SK, Ryymin PS, et al: MR angiography with three-dimensional time-of-flight and targeted maximum-intensity projection reconstructions in the follow-up of intracranial aneurysms embolized with Guglielmi detachable coils. AJNR Am J Neuroradiol 20:1470-1475, 1999.

20. Liang EY, Chan M, Hsiang JHK, et al: Detection and assessment of intracranial aneurysms: Value of CT angiography with shaded-surface display. AJR 165:1497-1502, 1995.

21. Ogawa T, Okudera T, Noguchi K, et al: Cerebral aneurysms: Evaluation with three-dimensional CT-angiography. AJNR 17:447-454, 1996.

22. Schwartz RB, Tice HM, Hooton SM, et al: Evaluation of cerebral aneurysms with helical CT: Correlation with conventional angiography and MR angiography. Radiology 192:717-722,1994.

23. Vieco PT, Shuman WP, Also from GF, et al: Detection of circle of Willis aneurysms in patients with acute subarachnoid hemorrhage: Comparison of CT angiography and digital subtraction angiography. AJR 165:425-430, 1995.

24. Kuszyk BS, Beauchamp NJ Jr., Fishman EK: Neurovascular applications of CT angiography. Semin Ultrasound CT MR 19:394-404, 1998.

25. Berland LL, Smith JK: Multidetector-array CT: Once again, technology creates new opportunities.Radiology 209:327-329, 1998.

26. Connor JJ, Wojak JC: Intracranial Arteriovenous Malformation: General Consideration. In: Connor JJ, Wojak JC, (eds): Interventional Neuroradiology­­Strategies and Practical Techniques. pp 227-239. Philadelphia, WB Saunders Company, 1999.

27. Edelman RR, Wentz KU, Mattle HP, et al: Intracerebral arteriovenous malformations: Evaluation with selective MR angiography and venography. Radiology 173:831-837, 1989.

28. Schwartz RB: Neuroradiological applications of spiral CT. Semin Ultrasound CT MRI 15:139-147, 1994.