The images of MR angiography were constructed based on the change of intensity of MR signals (time-of flight technique) or the change in the phase of blood flow (phase-contrast technique). In this paper, we describe the principles of those imaging techniques. We also describe the basic principles of gradient-echo technique, a fast imaging technique used for MRA. FLASH, Spoiled-GRASS, GRASS, and FISP, which are representative imaging techniques using gradient-echo technique, were referred to for their sequences for imaging.
Technical advances in magnetic resonance angiography (MRA) are now rapidly progressing. Its spatial resolusion, however, remains lower than that of a conventional angiography. Nevertheless, clinical use of MRA in diagnoses of many vascular lesions in the central nervous system is considered to be established. In this report, we discuss diagnostic methods and problems on MRA of the brain and the neck from our recent experiences. It should be stressed that vascular images obtained by MRA are not those of vessels proper but of blood streams under the influence of a flow velocity.
Most previous reports have explored the efficacy of magnetic resonance angiography (MRA) as a noninvasive screening method for the detection of unruptured intracranial aneurysms and occlusive vascular disease. The purpose of this study was to assess the accuracy of three-dimensional (3D) time-of-flight (TOF) MRA in the identification of ruptured aneurysms presenting subarachnoid hemorrhage (SAH) as well as unruptured aneurysms. Sixty-two consecutive patients with 36 ruptured and 32 unruptured aneurysms were examined. MR angiograms and conventional contrast angiograms were retrospectively reviewed with regard to the relation between the detectability on MRA and size of aneurysm, location, and influence of subarachnoid hematoma. The size of ruptured aneurysms was significantly larger than that of unruptured aneurysms. Twenty-seven of 29 ruptured aneurysms in acute stage (93%) was detected on MRA, whereas 29 of 36 unruptured aneurysms (81%) and none of 4 ruptured aneurysms in subacute stage (0%) were detectable. Individual vessel visualization and image quality of MRA in patients with acute SAH were satisfactory, although in patients with subacute SAH, MRA showed poor visualization of vessels due to T1-shortening of subacute surrounding hematoma and presence of delayed vasospasm. Main causes of failure to detect aneurysms on MRA included limitation of spatial resolution, limited projections obtainable with maximum-intensity-projection (MIP) techniques and subacute hematoma simulating flow signal on MIP reconstructions. Improvement of present limitation of MRA requires higher spatial resolution, higher contrast resolution and a more adequate reconstruction method. MRA was considered to be a safe and useful examination for intracranial aneurysms especially in patients with acute SAH.
To assess the efficacy of MR angiography (MRA) in detecting small asymptomatic cerebral aneurysms in clinical examination of the brain (CEB), we investigated the MRA data of 7,665 cases in our CEB between July 1990 and June 1993. A 1.5T system (Sigma Advantage; GE Medical Systems) was used in this study. Of the 7,665 cases, 410 aneurysms in 360 cases (4.7%) were found on MRA. These 360 cases included 152 males (3.8% of 3,956 male cases) and 208 females (5.6% of 3,709 female cases); females were slightly more prevalent than males. Aneurysms were diffusely distributed in all generations of adults, but more common in the fifties of both sexes than in other generations. Of 410 aneurysms, 209 (50.9%) originated from the IC-ophthalmic artery junction (IC-Oph), 85 (20.7%) from the IC-posterior communicating artery junction (ICPC), 57 (13.9%) from the middle cerebral artery (MCA), and 24 (5.9%) from the anterior communicating artery. IC-Oph aneurysms counted for half of all aneurysms. The size of aneurysms ranged from 1 mm to 14 mm. Two hundred and seventy-four aneurysms (66.8%) were less than 3 mm in diameter, 109 (26.6%) were 4-6 mm, 23 (5.6%) 7-9 mm, and 4 (0.9%) over 10 mm. Twenty-seven aneurysms 1 mm in diameter were the smallest found in our CEB and were distributed to IC-Oph in 17 cases, ICPC in 7, MCA in 2, and internal carotid artery bifurcation in 1. Conventional cerebral angiography was performed in 121 aneurysms in 107 cases, for whom we scheduled operation. One hundred and nine aneurysms in 96 cases (90.1%) were confirmed by angiography, while there were 12 false-positive (9.9%) and 6 false-negative cases. The presence of these false aneurysms was due to limited performance of the modern MRA system; insufficient spatial resolution or disappearance of small vessels. We conclude that it is necessary to read not only MRA, but axial source images of MRA carefully in order to detect aneurysms smaller than 3 mm and to decrease false cases as much as possible.
The usefulness of MR angiography (MRA) for the detection of the cerebral vasospasm after subarachnoid hemorrhage (SAH) was studied. Fourteen out of 69 patients of SAH between Jan. 1992 and Dec. 1993 underwent both CAG and MRA at the time of pre-spasm, spasm and post-spasm period. MRA was performed using an MRT-200FXII (Toshiba Corp., Tokyo. 1.5T). Most images were obtained using the 3D-TOF subtraction method. Cerebral arteries were divided into 4 portions, which consisted of A1, A2 and M1, M2. The angiographical findings of both CAG and MRA were studied at each divided area (such as M1 or M2) in the view of the presence or absence of vasospasm. Cerebral vasospasm on CAG and MRA were defined as arteries with a caliber less than 75% and 50% of the original size respectively. The compatibilities between the findings of MRA and CAG were 98% at the territory of ACA and 91% at MCA. The territory of 22% of the arteries was not visualized on MRA due to the artifact caused by aneurysmal clips. However, the total compatibility of MRA and CAG was 94%. The MRA is useful and will become the noninvasive diagnostic tool for the detection of the cerebral vasospasm.
A 46-year old man transferred to our hospital with complaints of severe sudden headache and repeated vomiting. CT scan revealed subarachnoid hemorrhage. Subarachnoid hemorrhage thickly covered the left Sylvian fissure and suprasellar cistern. MR angiography demonstrated a cerebral aneurysm, about 5 mm in diameter, at the left middle cerebral artery bifurcation. The patient underwent direct clipping of the aneurysm without conventional cerebral angiography. The postoperative course was uneventful. We suggest that conventional cerebral angiogram may be omitted with sufficient information from MR angiogram in the aneurysmal surgery.
Advantages and disadvantages of multiple MR angiographic sequences were investigated in 8 patients. Artifacts present on time of flight (TOF) images were due to substances with short T1, hemosiderin, surgical clips and fats. Phase contrast (PC) angiography allowed velocity resolution of vascular lesions and flow direction and had excellent background suppression. Gadopentetate dimeglumine enhancement increased the visualization of the vascular structures on both TOF and PC images. Two-dimensional (2D) TOF images were useful in slow flow such as dilated vessels and draining veins from AVMs. Three-dimensional (3D) TOF images were excellent in depicting small AVMs. Large AVMs were selected in the 2D TOF or 2D PC technique. AVMs with hematoma or surgical clips, and marked elevated flow had to be examined by 3D PC with the compatible velocity encoding provided with 2D PC. 2D PC images were useful in the assessment of the patency of major vascular structures.
Recent advances in Magnetic Resonance (MR) imaging technique have offered remarkable noninvasive diagnotic tool, MR angiography (MRA), to evaluate intracranial vascular lesion. MR angiography showed a good correlation with conventional angiography in diagnosing arteriosclerotic narrowing of intracranial arteries. In this paper, we review 3D time-of-flight MRA of the intracranial stenotic lesion and discuss potential pitfalls in flow evaluation by MRA. Because the contrast of the vessels comes from blood motion, artifactual signal loss can be caused by slight changes in normal blood streams such as eddy flow or decrease in flow velosity. Another signal loss can occur in a Maximum Intensity Projection (MIP) process. These effects lead to an apparent reduction in vessel diameter, an overestimation of stenosis and a loss of visualization of small or slow-flowing vessels. Arterial flow void on MRI, phase contrast MRA with low velosity encoding and original data before the MIP process may be useful for accurate diagnosis.
MR angiography (MRA) has been rapidly developed as a clinically useful modality. We present our experiences with MRA in the diagnosis and surgical treatment of Moyamoya disease. We have examined patients with Moyamoya disease using a 1.5 T MR unit (Signa, GE). The findings were analyzed in comparison with those of conventional cerebral angiography. A 0.5 T MR unit (Gyroscan, Phillips) was also used in some cases to compare MRA findings under different magnetic field strength. 3-D Time-of-Flight (3D-TOF) imaging was very sensitive in detecting steno-occlusive changes at the supraclinoid portion of the internal carotid artery and/or the origins of the anterior and middle cerebral arteries. There was, however, a tendency to overestimate the stenotic changes. In half the cases, it was difficult to visualize Moyamoya vessels clearly. 2D-TOF imaging detected the Moyamoya vessels better than 3D-TOF. MRI, which shows the Moyamoya vessels in the basal ganglia as flow void, is also helpful for diagnosis. In about 90% of the cases Moyamoya disease could be diagnosed by a combination of 3D-TOF, 2D-TOF and/or MRI without angiography. MRA at 0.5 T also detected the steno-occlusive lesions but its overestimation was more than that by 1.5T unit. Postoperative collateral formation from the external carotid artery to the brain was also observed on MRA by presaturation method. MRA can be used in the follow-up study at OPD and as a screening of Moyamoya disease.
To evaluate the contribution of magnetic resonance angiography (MRA) in the screening study of the extracranial carotid and vertebral arteries using the conventional head and neck coils, 500 consecutive MRAs of the cervical vessels were performed using 1.5 tesla magnetic resonance unit with circularly polarized head coil. The 5cm-thick imaging plane was placed in coronal fashion including both carotid and vertebral arteries. The imaging sequence was three-dimensional (3D) fast imaging with steady precession (FISP). In 10 patients with failed head coil examination, 10 patients with possible carotid and vertebral diseases and 10 volunteers, the extracranial carotid and vertebral arteries were examined with the Helmholtz neck coil. Both 3D- and 2D-FISP were performed in each case. The imaging plane was placed in oblique sagittal fashion. In 458 out of 500 cases (91.6%), the extracranial carotid and vertebral arteries were successfully depicted using head coil. In 20 patients with high shoulders, the carotid bifurcations were out of range of the head coil. In these cases, carotid bifurcations and the origins of the carotid and vertebral arteries were successfully revealed using a neck coil. To evaluate the stenotic lesions and tortuous vessels, 2D-FISP sequence seemed to be more suitable than 3D-FISP. Compared with conventional angiography, MRA caused overestimation of the degree of stenotic lesions. For screening examination of the extracranial carotid and vertebral arteries, most cases can be evaluated only with the conventional head coil. If depiction of the carotid bifurcation fails and the examination of carotids or vertebrals down to the aortic arch is needed, neck coil examination is required.
The authors have already reported surgical problems of poor-risk patients with hemorrhagic infratentorial arteriovenous malformations (AVMs). In the presents study, fifteen good-risk patients of infratentorial AVMs, whose levels of consciousness was 0 to 30 according to the Japan Coma Scale, were reviewed and the problems of the surgical treatment are reported. Of this series, 12 patients had a good result, and 3 patients a poor result. Giant AVM and AVM extending from the cerebellum to the brainstem showed an unsatisfactory surgical result. Two out of 3 poor results were due to massive postoperative hemorrhage, related to a residual nidus in the brainstem in one patient, and probably related to a sudden rise in intravascular pressure after removal of giant AVM in one. The present study suggests that presurgical multiple-staged embolization of nidus and a strict control of postoperative systemic blood pressure are important in order to reduce surgical risks of giant AVMs. Furthermore, stereotactic radiosurgery must be considered in combination with surgical treatment for an AVM involving the brainstem.
Forty-five carotid artery plaques (27 symptomatic, 18 asymptomatic) were obtained from carotid endarterectomies (25 unilateral, 10 bilateral) in 35 patients and were evaluated pathologically. We investigated the relationship of morphologic characteristics to the presence of cerebral ischemic symptoms and to the degree of stenosis associated with the carotid plaques. Simple fibrous plaques were found in 18 of the 45 plaques (40.0%) and were more frequently observed among the asymptomatic plaques (55.6%) than among the symptomatic plaques (29.6%). Ulceration was observed in 14 of the 45 plaques (31.1%) and was more frequent in plaques from symptomatic cases (44.4%) than in those from asymptomatic cases (11.1%). Six of the 12 plaques with ulceration showed intraplaque hemorrhage in the symptomatic cases. Eight of the 12 plaques with ulceration showed intraluminal thrombus in the symptomatic cases. Intraplaque hemorrhage was found in 17 of the 45 plaques (37.8%) and was more frequent in the plaques of symptomatic cases (48.1%) than those of asymptomatic cases (22.2%). Six of the 13 plaques with intraplaque hemorrhage showed ulceration in the symptomatic cases. Intraluminal thrombus was observed in 10 of the 45 plaques (26.7%) and more frequently in the plaques of symptomatic cases (37.0%) than in those of asymptomatic cases (11.1%). All plaques were divided into three broad degrees of stenosis groups (40-69%, 70-89%, 90-99%) angiographically. The incidence of simple fibrous plaques were most frequent in the most stenotic group, both when all plaques were considered and when only symptomatic plaques were examined. Although ulceration occurred more commonly in the 70-89% stenosis group, it was also observed in 4 plaques with relatively mild stenosis, and observed in only one plaque with more than 90% stenosis. Intraplaque hemorrhage occurred more commonly in the stenotic groups with more than 70% stenosis, whether all plaques or only symptomatic plaques were examined. These findings indicate that: 1) when stenosis of the carotid artery is severe, simple fibrous plaque may cause cerebral ischemia due to flow restriction, but a critical or hemodynamically significant internal carotid artery stenosis may be of a higher grade than previously appreciated when the lesion is a simple plaque, 2) ulceration of a carotid plaque is seen at a smaller plaque size and induces cerebral ischemia due to embolic phenomena, 3) since intraplaque hemorrhage may contribute to the acute progression of carotid stenosis, hemorrhage may cause hemodynamic failure. Intraplaque hemorrhage may also play an important role in ulceration to cause embolic phenomena.
The intracranial dissecting aneurysms (DA) and fusiform aneurysms (FA) of the vertebro-basilar system have been thought to be rare, but recently, reports about them have increased in number. However, how to decide therapy and estimate prognosis have been vague still now. We experienced 13 cases with DA or FA in the vertebro:basilar system from 1985 to now and 6 cases among them underwent follow-up angiography over 6 months. Six cases were male and 7 cases were female. The age of the cases ranged from 41 to 72 years old with a mean of 55.2 years old. The initial symptom was subarachnoid hemorrhage (SAH) in 9 cases and ischemic attack in 4 cases. DA or FA was involved in the right vertebral artery in 8 cases, in the left vertebral artery in 2 cases, and in the basilar artery in 3 cases. Angiographically, pearl and string sign or double lumen were demonstrated in 7 cases. In the other cases, fusiform shape or irregularly stenotic lumen were demonstrated. Four cases rebled during the acute stage and 3 cases died of rebleeding. One case had excellent recovery after emergent proximal clip occlusion of the right vertebral artery. Four cases without rebleeding had good outcomes. All 4 cases with ischemic attack had good recovery with conservative treatment. Follow-up angiography performed over 6 months after onset showed improvement of DA in 3 cases, dissapearance in 1 case, worsening in 1 case and no change in 1 case.