Clin Neuroradiol DOI 10.1007/s00062-015-0484-x
O R I G I N A L A RT I C L E
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative Non-Perimesencephalic Subarachnoid Hemorrhage J. M. Coutinho · R. H. Sacho · J. D. Schaafsma · R. Agid · T. Krings · I. Radovanovic · C. C. Matouk · D. J. Mikulis · D. M. Mandell
Received: 28 September 2015 / Accepted: 5 November 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose Standard magnetic resonance imaging (MRI) rarely identifies the cause of hemorrhage in patients with an angiogram-negative, non-perimesencephalic subarachnoid hemorrhage (SAH). Yet up to 10 % of these patients have recurrent hemorrhage. The aim of the study was to explore the potential role of high-resolution contrast-enhanced 3-Tesla vessel wall-MRI in patients with angiogram-negative SAH. Methods We performed intracranial vessel wall-MRI of the circle of Willis using a 3-Tesla scanner in consecutive patients presenting with a spontaneous, angiogram-negative,
D. M. Mandell () · J. M. Coutinho · R. H. Sacho · R. Agid · T. Krings · D. J. Mikulis Division of Neuroradiology, Department of Medical Imaging, Toronto Western Hospital, University Health Network and the University of Toronto, 399 Bathurst St, M5T 2S8 Toronto, ON, Canada e-mail: [email protected]
non-perimesencephalic SAH. Vessel wall-MRI included T1-, T2-, and gadolinium-enhanced T1-weighted two-dimensional black-blood sequences in multiple planes (voxel size 0.4 × 0.4 × 2.0 mm). Two neuroradiologists independently scored abnormalities of the arterial wall. Results In all, 11 patients (mean age 59 years) underwent vessel wall-MRI. A total of seven patients had vessel wall abnormalities despite normal catheter angiography. Two patients had focal abnormalities contiguous with the outer margin of the basilar artery wall for which we considered a differential of ruptured blood blister aneurysm, thrombosed aneurysm, and loculated extramural blood from elsewhere. Two patients had arterial wall enhancement involving multiple arteries, possibly secondary to SAH. Three patients had arterial wall enhancement at sites of dural penetration, remote from the SAH, likely related to age and atherosclerotic risk factors. Vessel wall-MRI did not alter patient management in this cohort. Conclusion Vessel wall-MRI showed abnormalities in seven patients with angiogram-negative SAH. These findings did not alter patient management, but the findings may be useful for other physicians who choose to perform vessel wall-MRI in this patient population.
J. D. Schaafsma Division of Neurology, Department of Medicine, University Health Network and the University of Toronto, Toronto, ON, Canada
Keywords Subarachnoid Hemorrhage · Intracranial aneurysm · Magnetic resonance imaging · High-resolution vessel wall magnetic resonance imaging
I. Radovanovic Division of Neurosurgery, Department of Surgery, University Health Network and the University of Toronto, Toronto, ON, Canada
Introduction
This work was presented in poster form at the 1st European Stroke Organization Congress in Glasgow, United Kingdom, 2015.
C. C. Matouk Departments of Diagnostic Radiology and Neurosurgery, Yale University School of Medicine, New Haven, CT, USA
The most common cause of nontraumatic subarachnoid hemorrhage (SAH) is cerebral aneurysm rupture. In 15 % of the patients with nontraumatic SAH, angiography shows
13
2
no aneurysm or other cause such as an arterial dissection, dural arteriovenous fistula, or vasculitis [1–4]. Among these the patients with angiogram-negative SAH, there is a subgroup who have a perimesencephalic distribution of blood, and this is associated with low risk of recurrent hemorrhage and excellent clinical outcome [5]. However, the patients with a non-perimesencephalic angiogramnegative SAH may develop hydrocephalus, vasospasm, and infarction; up to 10 % of these patients have recurrent hemorrhage, and approximately 20 % have poor clinical outcome [2, 6–8]. It is standard practice to perform a 1–2 week follow-up cerebral angiogram in patients with non-perimesencephalic angiogram-negative SAH. This delayed angiogram reveals an aneurysm in 7–14 % of the patients [9, 10]. False negative results on the initial angiogram may be due to thrombus formation within the aneurysm sac or vasospasm. Some centers also perform magnetic resonance imaging (MRI) of the brain and/or cervical spine which may identify a source of hemorrhage such as a brain tumor, cavernoma, or spinal vascular malformation [2–4]. However, the additional diagnostic yield of standard MRI is very low [11–14]. High-resolution 3-Tesla vessel wall magnetic resonance imaging (VW-MRI) has emerged as a technique that can demonstrate intracranial arterial abnormalities that are not apparent on standard MRI of the brain or luminal imaging (computed tomography (CT) angiography, MR angiography, or catheter angiography) [15–17]. VW-MRI can directly image a small focal thrombus and therefore has the potential to demonstrate: a small extravascular thrombus adjacent to rupture of a shallow sidewall (blood blister) aneurysm, an aneurysm sac that is completely filled with thrombus, or the intramural hematoma of a small arterial dissection. All of these pathologies can be difficult to detect angiographically. There is also preliminary evidence that VW-MRI demonstrates arterial wall enhancement at sites of recent aneurysm rupture [15]. Finally, VW-MRI can demonstrate arterial wall abnormality suggestive of vasculitis, which is a potential source of SAH [16, 18]. For example, preliminary evidence suggests that vasculitis typically manifests as a smooth and relatively concentric thickening and enhancement of the arterial wall, whereas atherosclerotic plaque is more often an eccentric abnormality. Despite its potential, there are no published reports of high-resolution VW-MRI in patients with non-perimesencephalic angiogram-negative SAH. The aim of this prospective pilot study was to explore the potential utility of high-resolution contrast-enhanced VWMRI in patients with angiogram-negative, non-perimesencephalic SAH.
13
J. M. Coutinho et al.
Methods Patients This was a prospective single-center pilot study. Between November 2012 and October 2014, we performed contrastenhanced intracranial VW-MRI in patients with an angiogram-negative, non-perimesencephalic SAH. We excluded patients with a perimesencephalic hemorrhage. The definition of a perimesencephalic hemorrhage was based on established criteria: [19] center of the hemorrhage located immediately anterior to the midbrain, with or without extension of blood to the anterior part of the ambient cistern or to the basal part of the Sylvian fissure; no extension to the anterior interhemispheric fissure and no extension to the lateral Sylvian fissure, except for a small amount of blood; and absence of frank intraventricular hemorrhage. We excluded patients with recent trauma that could explain the SAH. We also excluded patients with hemorrhage predominantly located in the sulci of the cerebral convexities rather than the basal cisterns, as the VW-MRI technique used for this study does not provide sufficient spatial coverage to evaluate the cerebral convexities. The study was approved by the Institutional Review Board of the University Health Network, Toronto. Routine Imaging All patients had CT angiography and catheter angiography at presentation. The catheter angiograms included injection of both internal carotid arteries, external carotid arteries, and vertebral arteries unless there was a specific impediment such as severe atherosclerosis. The catheter angiograms included three-dimensional rotational views of both internal carotids and the dominant vertebral artery. All patients had a second angiogram of the circle of Willis within 2 weeks using CT angiography, MR angiography, or catheter angiography. Other imaging, such as MRI of the brain and cervical spine or a third catheter angiogram, was done at the discretion of the treating physician. VW-MRI Protocol We performed VW-MRI using a Signa HDx 3.0-T scanner with an eight-channel head coil (GE Healthcare, Milwaukee, WI). The protocol included a time-of-flight MR angiogram of the circle of Willis, a T1-weighted black-blood vessel wall sequence (single inversion recovery-prepared two-dimensional fast spin echo acquisition with field of view = 22 × 22 cm, acquired matrix = 512 × 512; slice thickness = 2 mm; total slab thickness = 3–5 cm, TR/TI/TE = 2263/860/13 ms) before and after intravenous administration of gadolinium (with constant scan parameters),
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative Table 1 Baseline characteristics Patient Age (years) Sex Clinical presentation 1 51 F Headache, n/v, neck pain 2 35 F Headache, n/v 3 71 M Headache, n/v 4 66 F Headache, n/v 5 6
74 52
M F
Medical history Hypertension, hypothyroidism Unremarkable Unremarkable Addison’s disease, hypothyroidism, hypertension, diabetes Hypertension, colorectal cancer Unremarkable
3
Antiplatelets No No No No
WFNS grade 1 1 1 2
Headache, n/v No 2 Headache, n/v, neck pain, No 1 vertigo 7 49 F Headache, n/v Dyslipidemia No 1 8 58 F Headache, n/v, neck pain Unremarkable No 1 9 38 M Headache, n, neck pain Unremarkable No 1 10 83 M Headache, n/v, confusion Hypertension, coronary heart disease Aspirin 2 11 76 F Headache Myocardial infarction No 1 Headache was acute-onset in all patients HCF = hydrocephalus. IVH = intraventricular hemorrhage. n/v = nausea and vomiting
and a T2-weighted black-blood vessel wall sequence (twodimensional fast spin echo acquisition with field of view = 22 × 22 cm, acquired matrix 512 × 512; slice thickness = 2 mm; total slab thickness = 3–5 cm, TR/TE = 3250/89 ms). The vessel wall sequences were repeated to cover the circle of Willis in both the axial and coronal planes. We aimed to perform VW-MRI during the initial work-up. When this was not possible (due to patient condition or scanner schedule limitations), we performed VW-MRI when the patient returned for follow-up at our outpatient aneurysm clinic. Data Collection and Analysis We recorded age, sex, and medical history for each patient. We categorized SAH severity using the World Federation of Neurosurgical Societies (WFNS) Scale, and scored clinical outcome with the Glasgow Outcome Scale (GOS) during follow-up outpatient clinic visits. After the clinical data was acquired for all the patients in the study, two neuroradiologists reviewed all imaging (including VW-MRI) independently from each other. The readers categorized the technical quality of each VW-MRI examination as good, suboptimal but sufficient for diagnostic use, or nondiagnostic. Each reader recorded any abnormal findings seen on the MR angiogram and VW-MRI. We calculated Cohen’s kappa to measure interobserver agreement for quality of the study and VW-MRI findings on a patient-by-patient basis. We resolved discrepancies in the interpretation of the imaging by consensus. Results In all, 11 patients with an angiogram-negative non-perimesencephalic SAH underwent VW-MRI during the study period. The mean age was 59 years (range 35–83) and seven of the patients were women. A total of, eight patients had
HCF Yes Yes No No
IVH Yes Yes Yes Yes
Yes Yes
Yes Yes
No No Yes Yes No
Yes No No Yes No
a grade 1 (WFNS) SAH and three patients had a grade 2 SAH. Eight patients had intraventricular extension of the hemorrhage and six had acute hydrocephalus. Table 1 lists demographic and clinical details. Every patient had a CT angiogram and a catheter angiogram at presentation. In all, nine patients had a follow-up CT angiogram or catheter angiogram. One patient (patient 8) had an incidental extradural aneurysm of the right internal carotid artery. Another patient (patient 7) had mild ectasia of the right middle cerebral artery, but this was deemed an unlikely cause of the SAH due to the spatial distribution of the blood and stability of this ectasia at 3-month followup. One patient (patient 3) had vasospasm in the anterior and posterior circulation. A total of three patients underwent standard MRI of the brain and cervical spine, and none of these MRI exams identified a cause of hemorrhage. The median time from diagnosis of SAH to VW-MRI was 10 days (range 1–182 days; Table 2). Quality of the VW-MRI was rated as good in 9 patients, suboptimal in 2 patients due to mild head motion on some pulse sequences (patients 5 and 10), and poor in no patients. There was complete agreement between the raters on the quality of the imaging (K = 1.0). Interobserver agreement on VW-MRI findings was very good, with agreement in 10 of 11 studies (K = 0.81). There was disagreement on the interpretation of the VW-MRI findings 1 patient (patient 11), which was resolved by consensus upon re-examination. VW-MRI demonstrated a focal abnormality in two patients, both of whom were imaged between 1 and 2 weeks after SAH: One patient (patient 9) had a 2 mm well-defined round focus of intermediate signal on T1-weighted images and marked hypointensity on T2-weighted images in the prepontine cistern, contiguous with the outer margin of the mid basilar artery (Fig. 1). At the junction between the basilar artery wall and this 2 mm focus, there was an additional punctate focus which was markedly hyperintense on both
13
4
J. M. Coutinho et al.
Table 2 Intracranial vessel wall imaging findings and clinical outcome Patient VW-MRI tech- Interval SAH VW-MRI findings nical quality -VW-MRI (days) 1 Good 182 Mild arterial enhancement at site of dural penetration 2 Good 6 Unremarkable 3 Good 10 Focal abnormality distal basilar artery 4 Good 4 Smooth enhancement distal basilar artery and bilateral PCAs 5 Suboptimal 9 Mild arterial enhancement at site of dural penetration 6 Good 92 Unremarkable 7 Good 104 Unremarkable 8 Good 95 Unremarkable 9 Good 11 Focal abnormality mid basilar artery 10 Suboptimal 1 Mild arterial enhancement at site of dural penetration 11 Good 4 Minimal enhancement posterior wall basilar artery
sequences. These signal characteristics are most consistent with deoxygenated hemoglobin and extracellular methemoglobin, respectively [20]. The focal abnormality was visible against a background of more diffuse SAH elsewhere in the basal cisterns which was isointense-to-hyperintense on T2-weighted images and hypointense on T1-weighted images, and hyperattenuating on a CT scan performed the same day. Due to the VW-MRI appearance, the patient underwent repeat catheter angiography the same day, with magnified views of the basilar artery, but again there was no evidence of luminal irregularity or aneurysm. We repeated the VW-MRI 10 days later, which showed resolution of the focal abnormality. Another patient (patient 3) had a round focus of signal alteration adjacent to the posterior wall of the distal basilar artery. This focus was isointense on T1-weighted images and markedly hypointense on T2-weighted images (most consistent with deoxygenated hemoglobin) with no or minimal enhancement. This focus was visible against a background of more diffuse SAH elsewhere in the basal cisterns which was hyperintense on both T2- and T1-weighted images. At the same level as the focal abnormality, the SAH which was otherwise mainly ventral to the basilar artery extended focally around the posterior aspect of the basilar artery. There was associated severe narrowing of the distal third of the basilar artery and moderate narrowing of both posterior cerebral arteries, and there were small infarcts in the pons. We were not able to obtain follow-up VW-MRI for this second patient. VW-MRI demonstrated smooth arterial wall enhancement extending partway around the circumference of the distal basilar artery and proximal posterior cerebral arteries in 1 patient (patient 4), and similar but very mild arterial wall enhancement in another patient (patient 11). VW-MRI
13
Discharge destination Home
Duration of follow-up (months) 3
GOS 5
Home Home Home
2 2 NA
5 5 NA
Other hospital
NA
NA
Home Home Home Home Home
2 2 22 1 2
5 5 5 5 4
Home
NA
NA
was performed within 1 week of the SAH in both of these cases. Figure 2 shows representative images. VW-MRI demonstrated mild arterial wall enhancement extending partway around the circumference of the internal carotid and/or vertebral arteries around the site of dural penetration in three patients (patients 1, 5, and 10). Figure 3 shows representative images. All patients were treated with nimodipine during admission and all were alive at discharge. In all, 10 of 11 patients were discharged home, and 1 patient (patient 5) was transferred to a community hospital and subsequently lost to follow-up (Table 2). GOS scores were available for 8 of 11 patients, with a median duration of follow-up of 2 months (range 1–23 months). All GOS scores were five except for one patient with a score of four. There were no recurrent hemorrhages or other neurological events in this cohort of patients during follow-up. Discussion This study explored the potential role of high-resolution VW-MRI in patients with non-perimesencephalic, angiogram-negative SAH. In this cohort, the VW-MRI did not alter patient management in any of the 11 patients. However, 7 of 11 patients had VW-MRI abnormalities. Given a lack of literature on VW-MRI in patients with angiogramnegative SAH, we anticipate that our description and discussion of these findings will be useful for other clinicians who choose to perform VW-MRI in this patient population. VW-MRI demonstrated a focal abnormality contiguous with the outer margin of the basilar artery wall in two patients. One possible explanation for this abnormality is a thrombosed saccular aneurysm arising from the basilar sidewall or a basilar trunk perforator artery, with thrombus in
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative
5
Fig. 1 Three-dimensional surface rendering of a catheter angiogram, left vertebral artery injection (a) shows no evidence of aneurysm (patient 9). Axial nonenhanced computed tomography image (b) shows acute blood (arrows) in the prepontine cistern. High-resolution axial T2weighted vessel wall magnetic resonance imaging (VW-MRI) (c), performed 11 days after SAH) shows a 2 mm round region of marked hypointensity (white arrow) contiguous with the outer margin of the mid basilar artery (black arrow). Axial T1weighted VW-MRI (d) shows a punctate focus of hyperintensity (white arrow) at the junction between the basilar artery (dotted arrow) and the 2 mm region. Coronal T2-weighted VW-MRI (e) shows the 2 mm region of hypointensity contiguous with the outer margin of the basilar artery in a second plane. Axial T2-weighted VW-MRI 2 months later (f) shows resolution of the previous finding
the aneurysm lumen accounting for the focal region of signal abnormality on MRI. However, in one of these cases we were able to obtain short interval follow-up VW-MRI, and
this showed that the signal abnormality resolved over time, with no evidence of a residual underlying aneurysm sac, so this possibility seems unlikely. Another possible explana-
13
6
J. M. Coutinho et al.
Fig. 2 Axial thick slab reformat of a computed tomography (CT) angiogram (a) shows no evidence of aneurysm (patient 4). Nonenhanced axial CT image (b) shows acute blood around the posterior cerebral arteries. High-resolution axial T1-weighted vessel wall magnetic resonance imaging (VW-MRI) before (c) and after (d) intravenous injection of gadolinium show smooth enhancement of the arterial wall of the posterior cerebral arteries (VW-MRI performed 4 days after subarachnoid hemorrhage)
tion is a ruptured blood blister aneurysm (focal defect in the arterial wall covered with thin adventitia, resulting in just a shallow outpouching of the arterial wall on luminal imaging) [21, 22], with blood product outside the ruptured aneurysm wall accounting for the small focal signal abnormalities on VW-MRI. This mechanism could be consistent with the observed resolution of the focal signal abnormality over time. Arguing against this possibility, there was no angiographic evidence of a blister aneurysm even on careful catheter angiogram examination of the basilar artery with magnified views. Finally, it is possible that the focal signal
13
abnormality represents SAH that is loculated and incidentally located focally along the margin of the basilar artery, but is not indicative of the site of rupture. SAH is commonly loculated between arachnoidal membranes, although a focal loculation along the arterial wall, with signal characteristics that differ from the rest of the SAH, will likely at least cause pause when interpreting the VW-MRI. VW-MRI demonstrated mild arterial wall enhancement involving the internal carotid and/or vertebral arteries around the site of dural penetration (Fig. 3) in several patients. This VW-MRI finding has been observed in other
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative
7
Fig. 3 Coronal high-resolution T1-weighted vessel wall magnetic resonance imaging (VW-MRI) before (a) and after (b) intravenous injection of gadolinium show smooth enhancement (arrow) of the supraclinoid segment of the right internal carotid artery (patient 1, VW-MRI performed 182 days after subarachnoid hemorrhage)
contexts, and is attributable to the development of vasa vasorum associated with age and atherosclerotic risk factors [23]. Indeed, all of the patients with this finding were older than age 50 years and had known atherosclerotic risk factors. We therefore do not consider this finding a potential source of hemorrhage. However, when interpreting VWMRI, it is important to recognize this age-related phenomenon to avoid misdiagnosing the arterial wall enhancement as vasculitis. VW-MRI demonstrated long segments of smooth, partially circumferential arterial wall enhancement involving the distal basilar artery and proximal posterior cerebral arteries in two patients (Fig. 2). There was subarachnoid hemorrhage around these arteries, so it is possible that the arterial wall enhancement was an inflammatory response to the hemorrhage. Another possibility is an underlying vasculitis [18, 24]; however, the involvement of multiple vessels within a confined region of SAH is not typical for vasculitis, and these patients did not have any other clinical findings to suggest vasculitis. We have performed VW-MRI of the intracranial arterial wall, but it is possible that angiogram-negative SAH arises from venous hemorrhage. Indeed, the pathomechanism of perimesencephalic SAH is not proven, but there is evidence that the source is venous [2, 25]. Because the walls of normal intracranial veins are much thinner than arterial walls, intracranial venous VW-MRI may not be feasible. However, pathological alteration of a vein could make it thicker and visible, and successful wall imaging of larger veins in the leg has been reported [26]. Identifying the source of hemorrhage may also be less important when the source is venous as the prognosis is more likely favorable.
Several studies have assessed the utility of conventional brain MRI in patients with an angiogram-negative nonperimesencephalic SAH, and the diagnostic yield has generally been very low [11, 27]. However, VW-MRI differs considerably from conventional MRI so the results of these studies do not necessarily apply to VW-MRI. The VW-MRI technique we have used has a voxel size of 0.4 × 0.4 × 2.0 mm, and three-dimensional VW-MRI pulse sequences can achieve voxel size of 0.4 × 0.4 × 0.4 mm [28]. This compares with standard brain MRI pulse sequence which typically have voxel size in the order of 0.8 × 0.8 × 5.0 mm. Likely as important as high spatial resolution, VW-MRI employs black-blood pulse sequences which optimize visualization of the arterial wall relative to luminal blood and cerebrospinal fluid. In this study, we have use two-dimensional VW-MRI pulse sequences in multiple planes, and we were able to obtain high quality vessel wall images in short- and long-axis through arteries of interest. However, repeating sequences in multiple planes is time consuming and is not ideal for the assessment of arteries that curve. The relatively long scan duration and the necessity for patient compliance to avoid motion artifact can be problematic in patients with recent SAH, and may prevent sicker patients from undergoing VW-MRI unless they are intubated. Three-dimensional VW-MRI sequences can partly overcome this limitation, since the scanning time is substantially reduced with this technique [28]. The clinical characteristics of the patients in our study are mainly similar to previous cohorts of patients with an angiogram-negative nonperimesencephalic SAH, with a mildly higher incidence of hydrocephalus in our study [2, 6, 29].
13
8
The time between SAH and VW-MRI was longer than 3 months for four patients in our study. This was the result of logistic issues, but a long delay is undesirable as the overall goal is to identify the cause of hemorrhage early enough to prevent recurrent hemorrhage. As well, this delay may have reduced our sensitivity for detecting a cause of hemorrhage as some conditions such as an intracranial arterial dissection can spontaneously heal [30]. However, if VWMRI is performed too early, a large volume of blood in the subarachnoid space may confound identification of arterial wall abnormality on VW-MRI. Based on experience in the study, VW-MRI at 14 days after SAH may provide a reasonable balance between clearing of most blood on T1- and T2-weighetd VW-MRI and imaging early enough to potentially treat a lesion before recurrent hemorrhage. Conclusions To summarize, we explored the utility of high-resolution VW-VRI in the patients with non-perimesencephalic, angiogram-negative SAH. The VW-MRI findings did not alter patient management in our cohort. However, 7 of 11 patients had VW-MRI abnormalities, and given a lack of literature on VW-MRI in patients with angiogram-negative SAH, we hope that our description and discussion of our findings will be useful for other clinicians who perform high-resolution VW-MRI in this patient population in a research or clinical setting. Larger studies are likely warranted to further evaluate the diagnostic yield of VW-MRI in patients with nonperimesencephalic, angiogram-negative SAH. Financial Support JMC is supported by the Netherlands Organisation for Scientific Research (NWO), Thrombosis Foundation Holland, Dutch Heart Foundation, and the Remmert Adriaan Laan Fonds. Conflict of Interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
References 1. van Gijn J, Rinkel GJ. Subarachnoid haemorrhage: diagnosis, causes and management. Brain. 2001;124:249–78. 2. Boswell S, Thorell W, Gogela S, Lyden E, Surdell D. Angiogramnegative subarachnoid hemorrhage: outcomes data and review of the literature. J Stroke Cerebrovasc Dis. 2013;22:750–7. 3. Aviv RI, Shad A, Tomlinson G, Niemann D, Teddy PJ, Molyneux AJ, Byrne JV. Cervical dural arteriovenous fistulae manifesting as subarachnoid hemorrhage: report of two cases and literature review. AJNR Am J Neuroradiol. 2004;25:854–8. 4. Nicastro N, Schnider A, Leemann B. Anaplastic medullary ependymoma presenting as subarachnoid hemorrhage. Case Rep Neurol Med. 2013;2013:701820. 5. van Gijn J, Kerr RS, Rinkel GJ. Subarachnoid haemorrhage. Lancet. 2007;369:306–18.
13
J. M. Coutinho et al. 6. Konczalla J, Schuss P, Platz J, Vatter H, Seifert V, Guresir E. Clinical outcome and prognostic factors of patients with angiogramnegative and non-perimesencephalic subarachnoid hemorrhage: benign prognosis like perimesencephalic SAH or same risk as aneurysmal SAH? Neurosurg Rev. 2015;38:121–7. 7. Maslehaty H, Barth H, Petridis AK, Doukas A, Maximilian Mehdorn H. Special features of subarachnoid hemorrhage of unknown origin: a review of a series of 179 cases. Neurol Res. 2012;34:91–7. 8. Elhadi AM, Zabramski JM, Almefty KK, Mendes GA, Nakaji P, McDougall CG, Albuquerque FC, Preul MC, Spetzler RF. Spontaneous subarachnoid hemorrhage of unknown origin: hospital course and long-term clinical and angiographic follow-up. J Neurosurg. 2014:1–8. 9. Bakker NA, Groen RJ, Foumani M, Uyttenboogaart M, Eshghi OS, Metzemaekers JD, Lammers N, Luijckx GJ, Van Dijk JM. Repeat digital subtraction angiography after a negative baseline assessment in nonperimesencephalic subarachnoid hemorrhage: a pooled data meta-analysis. J Neurosurg. 2014;120:99–103. 10. Agid R, Andersson T, Almqvist H, Willinsky RA, Lee SK, terBrugge KG, Farb RI, Söderman M. Negative CT angiography findings in patients with spontaneous subarachnoid hemorrhage: when is digital subtraction angiography still needed? AJNR Am J Neuroradiol. 2010;31:696–705. 11. Woodfield J, Rane N, Cudlip S, Byrne JV. Value of delayed MRI in angiogram-negative subarachnoid haemorrhage. Clin Radiol. 2014;69:350–6. 12. Andaluz N, Zuccarello M. Yield of further diagnostic work-up of cryptogenic subarachnoid hemorrhage based on bleeding patterns on computed tomographic scans. Neurosurgery. 2008;62:1040–6. 13. Rogg JM, Smeaton S, Doberstein C, Goldstein JH, Tung GA, Haas RA. Assessment of the value of MR imaging for examining patients with angiographically negative subarachnoid hemorrhage. AJR Am J Roentgenol. 1999;172:201–6. 14. Topcuoglu MA, Ogilvy CS, Carter BS, Buonanno FS, Koroshetz WJ, Singhal AB. Subarachnoid hemorrhage without evident cause on initial angiography studies: diagnostic yield of subsequent angiography and other neuroimaging tests. J Neurosurg. 2003;98:1235–40. 15. Matouk CC, Mandell DM, Gunel M, Bulsara KR, Malhotra A, Hebert R, Johnson MH, Mikulis DJ, Minja FJ. Vessel wall magnetic resonance imaging identifies the site of rupture in patients with multiple intracranial aneurysms: proof of principle. Neurosurgery. 2013;72:492–6. 16. Mandell DM, Matouk CC, Farb RI, Krings T, Agid R, terBrugge K, Willinsky RA, Swartz RH, Silver FL, Mikulis DJ. Vessel wall MRI to differentiate between reversible cerebral vasoconstriction syndrome and central nervous system vasculitis: preliminary results. Stroke. 2012;43:860–2. 17. Power S, Matouk C, Casaubon LK, Silver FL, Krings T, Mikulis DJ, Mandell DM. Vessel wall magnetic resonance imaging in acute ischemic stroke: effects of embolism and mechanical thrombectomy on the arterial wall. Stroke. 2014;45:2330–4. 18. Obusez EC, Hui F, Hajj-Ali RA, Cerejo R, Calabrese LH, Hammad T, Jones SE. High-resolution MRI vessel wall imaging: spatial and temporal patterns of reversible cerebral vasoconstriction syndrome and central nervous system vasculitis. AJNR Am J Neuroradiol. 2014;35:1527–32. 19. Rinkel GJ, Wijdicks EF, Hasan D, Kienstra GE, Franke CL, Hageman LM, Vermeulen M, van Gijn J. Outcome in patients with subarachnoid haemorrhage and negative angiography according to pattern of haemorrhage on computed tomography. Lancet. 1991;338:964–8. 20. Bradley WG Jr. MR appearance of hemorrhage in the brain. Radiology. 1993;189:15–26.
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative 21. Horie N, Morikawa M, Fukuda S, Hayashi K, Suyama K, Nagata I. Detection of blood blister-like aneurysm and intramural hematoma with high-resolution magnetic resonance imaging. J Neurosurg. 2011;115:1206–9. 22. Gonzalez AM, Narata AP, Yilmaz H, Bijlenga P, Radovanovic I, Schaller K, Lovblad KO, Pereira VM. Blood blister-like aneurysms: single center experience and systematic literature review. Eur J Radiol. 2014;83:197–205. 23. Aoki S, Shirouzu I, Sasaki Y, Okubo T, Hayashi N, Machida T, Hoshi E, Suzuki K, Funada N, Araki T. Enhancement of the intracranial arterial wall at MR imaging: relationship to cerebral atherosclerosis. Radiology. 1995;194:477–81. 24. Mossa-Basha M, Hwang WD, De Havenon A, Hippe D, Balu N, Becker KJ, Tirschwell DT, Hatsukami T, Anzai Y, Yuan C. Multicontrast high-resolution vessel wall magnetic resonance imaging and its value in differentiating intracranial vasculopathic processes. Stroke. 2015;46:1567–73. 25. van der Schaaf IC, Velthuis BK, Gouw A, Rinkel GJ. Venous drainage in perimesencephalic hemorrhage. Stroke. 2004;35:1614–8. 26. Ambrose N, Pierce IT, Gatehouse PD, Haskard DO, Firmin DN. Magnetic resonance imaging of vein wall thickness in patients with Behcet’s syndrome. Clin Exp Rheumatol. 2014;32:S99–102.
9
27. Maslehaty H, Petridis AK, Barth H, Mehdorn HM. Diagnostic value of magnetic resonance imaging in perimesencephalic and nonperimesencephalic subarachnoid hemorrhage of unknown origin. J Neurosurg. 2011;114:1003–7. 28. Qiao Y, Steinman DA, Qin Q, Etesami M, Schar M, Astor BC, Wasserman BA. Intracranial arterial wall imaging using threedimensional high isotropic resolution black blood MRI at 3.0 T. J Magn Reson Imaging. 2011;34:22–30. 29. Konczalla J, Platz J, Schuss P, Vatter H, Seifert V, Guresir E. Non-aneurysmal non-traumatic subarachnoid hemorrhage: patient characteristics, clinical outcome and prognostic factors based on a single-center experience in 125 patients. BMC Neurol. 2014;14:140. 30. Wang Y, Lou X, Li Y, Sui B, Sun S, Li C, Jiang P, Siddiqui A, Yang X. Imaging investigation of intracranial arterial dissecting aneurysms by using 3 T high-resolution MRI and DSA: from the interventional neuroradiologists’ view. Acta Neurochir (Wien). 2014;156:515–25.
13
O R I G I N A L A RT I C L E
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative Non-Perimesencephalic Subarachnoid Hemorrhage J. M. Coutinho · R. H. Sacho · J. D. Schaafsma · R. Agid · T. Krings · I. Radovanovic · C. C. Matouk · D. J. Mikulis · D. M. Mandell
Received: 28 September 2015 / Accepted: 5 November 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose Standard magnetic resonance imaging (MRI) rarely identifies the cause of hemorrhage in patients with an angiogram-negative, non-perimesencephalic subarachnoid hemorrhage (SAH). Yet up to 10 % of these patients have recurrent hemorrhage. The aim of the study was to explore the potential role of high-resolution contrast-enhanced 3-Tesla vessel wall-MRI in patients with angiogram-negative SAH. Methods We performed intracranial vessel wall-MRI of the circle of Willis using a 3-Tesla scanner in consecutive patients presenting with a spontaneous, angiogram-negative,
D. M. Mandell () · J. M. Coutinho · R. H. Sacho · R. Agid · T. Krings · D. J. Mikulis Division of Neuroradiology, Department of Medical Imaging, Toronto Western Hospital, University Health Network and the University of Toronto, 399 Bathurst St, M5T 2S8 Toronto, ON, Canada e-mail: [email protected]
non-perimesencephalic SAH. Vessel wall-MRI included T1-, T2-, and gadolinium-enhanced T1-weighted two-dimensional black-blood sequences in multiple planes (voxel size 0.4 × 0.4 × 2.0 mm). Two neuroradiologists independently scored abnormalities of the arterial wall. Results In all, 11 patients (mean age 59 years) underwent vessel wall-MRI. A total of seven patients had vessel wall abnormalities despite normal catheter angiography. Two patients had focal abnormalities contiguous with the outer margin of the basilar artery wall for which we considered a differential of ruptured blood blister aneurysm, thrombosed aneurysm, and loculated extramural blood from elsewhere. Two patients had arterial wall enhancement involving multiple arteries, possibly secondary to SAH. Three patients had arterial wall enhancement at sites of dural penetration, remote from the SAH, likely related to age and atherosclerotic risk factors. Vessel wall-MRI did not alter patient management in this cohort. Conclusion Vessel wall-MRI showed abnormalities in seven patients with angiogram-negative SAH. These findings did not alter patient management, but the findings may be useful for other physicians who choose to perform vessel wall-MRI in this patient population.
J. D. Schaafsma Division of Neurology, Department of Medicine, University Health Network and the University of Toronto, Toronto, ON, Canada
Keywords Subarachnoid Hemorrhage · Intracranial aneurysm · Magnetic resonance imaging · High-resolution vessel wall magnetic resonance imaging
I. Radovanovic Division of Neurosurgery, Department of Surgery, University Health Network and the University of Toronto, Toronto, ON, Canada
Introduction
This work was presented in poster form at the 1st European Stroke Organization Congress in Glasgow, United Kingdom, 2015.
C. C. Matouk Departments of Diagnostic Radiology and Neurosurgery, Yale University School of Medicine, New Haven, CT, USA
The most common cause of nontraumatic subarachnoid hemorrhage (SAH) is cerebral aneurysm rupture. In 15 % of the patients with nontraumatic SAH, angiography shows
13
2
no aneurysm or other cause such as an arterial dissection, dural arteriovenous fistula, or vasculitis [1–4]. Among these the patients with angiogram-negative SAH, there is a subgroup who have a perimesencephalic distribution of blood, and this is associated with low risk of recurrent hemorrhage and excellent clinical outcome [5]. However, the patients with a non-perimesencephalic angiogramnegative SAH may develop hydrocephalus, vasospasm, and infarction; up to 10 % of these patients have recurrent hemorrhage, and approximately 20 % have poor clinical outcome [2, 6–8]. It is standard practice to perform a 1–2 week follow-up cerebral angiogram in patients with non-perimesencephalic angiogram-negative SAH. This delayed angiogram reveals an aneurysm in 7–14 % of the patients [9, 10]. False negative results on the initial angiogram may be due to thrombus formation within the aneurysm sac or vasospasm. Some centers also perform magnetic resonance imaging (MRI) of the brain and/or cervical spine which may identify a source of hemorrhage such as a brain tumor, cavernoma, or spinal vascular malformation [2–4]. However, the additional diagnostic yield of standard MRI is very low [11–14]. High-resolution 3-Tesla vessel wall magnetic resonance imaging (VW-MRI) has emerged as a technique that can demonstrate intracranial arterial abnormalities that are not apparent on standard MRI of the brain or luminal imaging (computed tomography (CT) angiography, MR angiography, or catheter angiography) [15–17]. VW-MRI can directly image a small focal thrombus and therefore has the potential to demonstrate: a small extravascular thrombus adjacent to rupture of a shallow sidewall (blood blister) aneurysm, an aneurysm sac that is completely filled with thrombus, or the intramural hematoma of a small arterial dissection. All of these pathologies can be difficult to detect angiographically. There is also preliminary evidence that VW-MRI demonstrates arterial wall enhancement at sites of recent aneurysm rupture [15]. Finally, VW-MRI can demonstrate arterial wall abnormality suggestive of vasculitis, which is a potential source of SAH [16, 18]. For example, preliminary evidence suggests that vasculitis typically manifests as a smooth and relatively concentric thickening and enhancement of the arterial wall, whereas atherosclerotic plaque is more often an eccentric abnormality. Despite its potential, there are no published reports of high-resolution VW-MRI in patients with non-perimesencephalic angiogram-negative SAH. The aim of this prospective pilot study was to explore the potential utility of high-resolution contrast-enhanced VWMRI in patients with angiogram-negative, non-perimesencephalic SAH.
13
J. M. Coutinho et al.
Methods Patients This was a prospective single-center pilot study. Between November 2012 and October 2014, we performed contrastenhanced intracranial VW-MRI in patients with an angiogram-negative, non-perimesencephalic SAH. We excluded patients with a perimesencephalic hemorrhage. The definition of a perimesencephalic hemorrhage was based on established criteria: [19] center of the hemorrhage located immediately anterior to the midbrain, with or without extension of blood to the anterior part of the ambient cistern or to the basal part of the Sylvian fissure; no extension to the anterior interhemispheric fissure and no extension to the lateral Sylvian fissure, except for a small amount of blood; and absence of frank intraventricular hemorrhage. We excluded patients with recent trauma that could explain the SAH. We also excluded patients with hemorrhage predominantly located in the sulci of the cerebral convexities rather than the basal cisterns, as the VW-MRI technique used for this study does not provide sufficient spatial coverage to evaluate the cerebral convexities. The study was approved by the Institutional Review Board of the University Health Network, Toronto. Routine Imaging All patients had CT angiography and catheter angiography at presentation. The catheter angiograms included injection of both internal carotid arteries, external carotid arteries, and vertebral arteries unless there was a specific impediment such as severe atherosclerosis. The catheter angiograms included three-dimensional rotational views of both internal carotids and the dominant vertebral artery. All patients had a second angiogram of the circle of Willis within 2 weeks using CT angiography, MR angiography, or catheter angiography. Other imaging, such as MRI of the brain and cervical spine or a third catheter angiogram, was done at the discretion of the treating physician. VW-MRI Protocol We performed VW-MRI using a Signa HDx 3.0-T scanner with an eight-channel head coil (GE Healthcare, Milwaukee, WI). The protocol included a time-of-flight MR angiogram of the circle of Willis, a T1-weighted black-blood vessel wall sequence (single inversion recovery-prepared two-dimensional fast spin echo acquisition with field of view = 22 × 22 cm, acquired matrix = 512 × 512; slice thickness = 2 mm; total slab thickness = 3–5 cm, TR/TI/TE = 2263/860/13 ms) before and after intravenous administration of gadolinium (with constant scan parameters),
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative Table 1 Baseline characteristics Patient Age (years) Sex Clinical presentation 1 51 F Headache, n/v, neck pain 2 35 F Headache, n/v 3 71 M Headache, n/v 4 66 F Headache, n/v 5 6
74 52
M F
Medical history Hypertension, hypothyroidism Unremarkable Unremarkable Addison’s disease, hypothyroidism, hypertension, diabetes Hypertension, colorectal cancer Unremarkable
3
Antiplatelets No No No No
WFNS grade 1 1 1 2
Headache, n/v No 2 Headache, n/v, neck pain, No 1 vertigo 7 49 F Headache, n/v Dyslipidemia No 1 8 58 F Headache, n/v, neck pain Unremarkable No 1 9 38 M Headache, n, neck pain Unremarkable No 1 10 83 M Headache, n/v, confusion Hypertension, coronary heart disease Aspirin 2 11 76 F Headache Myocardial infarction No 1 Headache was acute-onset in all patients HCF = hydrocephalus. IVH = intraventricular hemorrhage. n/v = nausea and vomiting
and a T2-weighted black-blood vessel wall sequence (twodimensional fast spin echo acquisition with field of view = 22 × 22 cm, acquired matrix 512 × 512; slice thickness = 2 mm; total slab thickness = 3–5 cm, TR/TE = 3250/89 ms). The vessel wall sequences were repeated to cover the circle of Willis in both the axial and coronal planes. We aimed to perform VW-MRI during the initial work-up. When this was not possible (due to patient condition or scanner schedule limitations), we performed VW-MRI when the patient returned for follow-up at our outpatient aneurysm clinic. Data Collection and Analysis We recorded age, sex, and medical history for each patient. We categorized SAH severity using the World Federation of Neurosurgical Societies (WFNS) Scale, and scored clinical outcome with the Glasgow Outcome Scale (GOS) during follow-up outpatient clinic visits. After the clinical data was acquired for all the patients in the study, two neuroradiologists reviewed all imaging (including VW-MRI) independently from each other. The readers categorized the technical quality of each VW-MRI examination as good, suboptimal but sufficient for diagnostic use, or nondiagnostic. Each reader recorded any abnormal findings seen on the MR angiogram and VW-MRI. We calculated Cohen’s kappa to measure interobserver agreement for quality of the study and VW-MRI findings on a patient-by-patient basis. We resolved discrepancies in the interpretation of the imaging by consensus. Results In all, 11 patients with an angiogram-negative non-perimesencephalic SAH underwent VW-MRI during the study period. The mean age was 59 years (range 35–83) and seven of the patients were women. A total of, eight patients had
HCF Yes Yes No No
IVH Yes Yes Yes Yes
Yes Yes
Yes Yes
No No Yes Yes No
Yes No No Yes No
a grade 1 (WFNS) SAH and three patients had a grade 2 SAH. Eight patients had intraventricular extension of the hemorrhage and six had acute hydrocephalus. Table 1 lists demographic and clinical details. Every patient had a CT angiogram and a catheter angiogram at presentation. In all, nine patients had a follow-up CT angiogram or catheter angiogram. One patient (patient 8) had an incidental extradural aneurysm of the right internal carotid artery. Another patient (patient 7) had mild ectasia of the right middle cerebral artery, but this was deemed an unlikely cause of the SAH due to the spatial distribution of the blood and stability of this ectasia at 3-month followup. One patient (patient 3) had vasospasm in the anterior and posterior circulation. A total of three patients underwent standard MRI of the brain and cervical spine, and none of these MRI exams identified a cause of hemorrhage. The median time from diagnosis of SAH to VW-MRI was 10 days (range 1–182 days; Table 2). Quality of the VW-MRI was rated as good in 9 patients, suboptimal in 2 patients due to mild head motion on some pulse sequences (patients 5 and 10), and poor in no patients. There was complete agreement between the raters on the quality of the imaging (K = 1.0). Interobserver agreement on VW-MRI findings was very good, with agreement in 10 of 11 studies (K = 0.81). There was disagreement on the interpretation of the VW-MRI findings 1 patient (patient 11), which was resolved by consensus upon re-examination. VW-MRI demonstrated a focal abnormality in two patients, both of whom were imaged between 1 and 2 weeks after SAH: One patient (patient 9) had a 2 mm well-defined round focus of intermediate signal on T1-weighted images and marked hypointensity on T2-weighted images in the prepontine cistern, contiguous with the outer margin of the mid basilar artery (Fig. 1). At the junction between the basilar artery wall and this 2 mm focus, there was an additional punctate focus which was markedly hyperintense on both
13
4
J. M. Coutinho et al.
Table 2 Intracranial vessel wall imaging findings and clinical outcome Patient VW-MRI tech- Interval SAH VW-MRI findings nical quality -VW-MRI (days) 1 Good 182 Mild arterial enhancement at site of dural penetration 2 Good 6 Unremarkable 3 Good 10 Focal abnormality distal basilar artery 4 Good 4 Smooth enhancement distal basilar artery and bilateral PCAs 5 Suboptimal 9 Mild arterial enhancement at site of dural penetration 6 Good 92 Unremarkable 7 Good 104 Unremarkable 8 Good 95 Unremarkable 9 Good 11 Focal abnormality mid basilar artery 10 Suboptimal 1 Mild arterial enhancement at site of dural penetration 11 Good 4 Minimal enhancement posterior wall basilar artery
sequences. These signal characteristics are most consistent with deoxygenated hemoglobin and extracellular methemoglobin, respectively [20]. The focal abnormality was visible against a background of more diffuse SAH elsewhere in the basal cisterns which was isointense-to-hyperintense on T2-weighted images and hypointense on T1-weighted images, and hyperattenuating on a CT scan performed the same day. Due to the VW-MRI appearance, the patient underwent repeat catheter angiography the same day, with magnified views of the basilar artery, but again there was no evidence of luminal irregularity or aneurysm. We repeated the VW-MRI 10 days later, which showed resolution of the focal abnormality. Another patient (patient 3) had a round focus of signal alteration adjacent to the posterior wall of the distal basilar artery. This focus was isointense on T1-weighted images and markedly hypointense on T2-weighted images (most consistent with deoxygenated hemoglobin) with no or minimal enhancement. This focus was visible against a background of more diffuse SAH elsewhere in the basal cisterns which was hyperintense on both T2- and T1-weighted images. At the same level as the focal abnormality, the SAH which was otherwise mainly ventral to the basilar artery extended focally around the posterior aspect of the basilar artery. There was associated severe narrowing of the distal third of the basilar artery and moderate narrowing of both posterior cerebral arteries, and there were small infarcts in the pons. We were not able to obtain follow-up VW-MRI for this second patient. VW-MRI demonstrated smooth arterial wall enhancement extending partway around the circumference of the distal basilar artery and proximal posterior cerebral arteries in 1 patient (patient 4), and similar but very mild arterial wall enhancement in another patient (patient 11). VW-MRI
13
Discharge destination Home
Duration of follow-up (months) 3
GOS 5
Home Home Home
2 2 NA
5 5 NA
Other hospital
NA
NA
Home Home Home Home Home
2 2 22 1 2
5 5 5 5 4
Home
NA
NA
was performed within 1 week of the SAH in both of these cases. Figure 2 shows representative images. VW-MRI demonstrated mild arterial wall enhancement extending partway around the circumference of the internal carotid and/or vertebral arteries around the site of dural penetration in three patients (patients 1, 5, and 10). Figure 3 shows representative images. All patients were treated with nimodipine during admission and all were alive at discharge. In all, 10 of 11 patients were discharged home, and 1 patient (patient 5) was transferred to a community hospital and subsequently lost to follow-up (Table 2). GOS scores were available for 8 of 11 patients, with a median duration of follow-up of 2 months (range 1–23 months). All GOS scores were five except for one patient with a score of four. There were no recurrent hemorrhages or other neurological events in this cohort of patients during follow-up. Discussion This study explored the potential role of high-resolution VW-MRI in patients with non-perimesencephalic, angiogram-negative SAH. In this cohort, the VW-MRI did not alter patient management in any of the 11 patients. However, 7 of 11 patients had VW-MRI abnormalities. Given a lack of literature on VW-MRI in patients with angiogramnegative SAH, we anticipate that our description and discussion of these findings will be useful for other clinicians who choose to perform VW-MRI in this patient population. VW-MRI demonstrated a focal abnormality contiguous with the outer margin of the basilar artery wall in two patients. One possible explanation for this abnormality is a thrombosed saccular aneurysm arising from the basilar sidewall or a basilar trunk perforator artery, with thrombus in
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative
5
Fig. 1 Three-dimensional surface rendering of a catheter angiogram, left vertebral artery injection (a) shows no evidence of aneurysm (patient 9). Axial nonenhanced computed tomography image (b) shows acute blood (arrows) in the prepontine cistern. High-resolution axial T2weighted vessel wall magnetic resonance imaging (VW-MRI) (c), performed 11 days after SAH) shows a 2 mm round region of marked hypointensity (white arrow) contiguous with the outer margin of the mid basilar artery (black arrow). Axial T1weighted VW-MRI (d) shows a punctate focus of hyperintensity (white arrow) at the junction between the basilar artery (dotted arrow) and the 2 mm region. Coronal T2-weighted VW-MRI (e) shows the 2 mm region of hypointensity contiguous with the outer margin of the basilar artery in a second plane. Axial T2-weighted VW-MRI 2 months later (f) shows resolution of the previous finding
the aneurysm lumen accounting for the focal region of signal abnormality on MRI. However, in one of these cases we were able to obtain short interval follow-up VW-MRI, and
this showed that the signal abnormality resolved over time, with no evidence of a residual underlying aneurysm sac, so this possibility seems unlikely. Another possible explana-
13
6
J. M. Coutinho et al.
Fig. 2 Axial thick slab reformat of a computed tomography (CT) angiogram (a) shows no evidence of aneurysm (patient 4). Nonenhanced axial CT image (b) shows acute blood around the posterior cerebral arteries. High-resolution axial T1-weighted vessel wall magnetic resonance imaging (VW-MRI) before (c) and after (d) intravenous injection of gadolinium show smooth enhancement of the arterial wall of the posterior cerebral arteries (VW-MRI performed 4 days after subarachnoid hemorrhage)
tion is a ruptured blood blister aneurysm (focal defect in the arterial wall covered with thin adventitia, resulting in just a shallow outpouching of the arterial wall on luminal imaging) [21, 22], with blood product outside the ruptured aneurysm wall accounting for the small focal signal abnormalities on VW-MRI. This mechanism could be consistent with the observed resolution of the focal signal abnormality over time. Arguing against this possibility, there was no angiographic evidence of a blister aneurysm even on careful catheter angiogram examination of the basilar artery with magnified views. Finally, it is possible that the focal signal
13
abnormality represents SAH that is loculated and incidentally located focally along the margin of the basilar artery, but is not indicative of the site of rupture. SAH is commonly loculated between arachnoidal membranes, although a focal loculation along the arterial wall, with signal characteristics that differ from the rest of the SAH, will likely at least cause pause when interpreting the VW-MRI. VW-MRI demonstrated mild arterial wall enhancement involving the internal carotid and/or vertebral arteries around the site of dural penetration (Fig. 3) in several patients. This VW-MRI finding has been observed in other
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative
7
Fig. 3 Coronal high-resolution T1-weighted vessel wall magnetic resonance imaging (VW-MRI) before (a) and after (b) intravenous injection of gadolinium show smooth enhancement (arrow) of the supraclinoid segment of the right internal carotid artery (patient 1, VW-MRI performed 182 days after subarachnoid hemorrhage)
contexts, and is attributable to the development of vasa vasorum associated with age and atherosclerotic risk factors [23]. Indeed, all of the patients with this finding were older than age 50 years and had known atherosclerotic risk factors. We therefore do not consider this finding a potential source of hemorrhage. However, when interpreting VWMRI, it is important to recognize this age-related phenomenon to avoid misdiagnosing the arterial wall enhancement as vasculitis. VW-MRI demonstrated long segments of smooth, partially circumferential arterial wall enhancement involving the distal basilar artery and proximal posterior cerebral arteries in two patients (Fig. 2). There was subarachnoid hemorrhage around these arteries, so it is possible that the arterial wall enhancement was an inflammatory response to the hemorrhage. Another possibility is an underlying vasculitis [18, 24]; however, the involvement of multiple vessels within a confined region of SAH is not typical for vasculitis, and these patients did not have any other clinical findings to suggest vasculitis. We have performed VW-MRI of the intracranial arterial wall, but it is possible that angiogram-negative SAH arises from venous hemorrhage. Indeed, the pathomechanism of perimesencephalic SAH is not proven, but there is evidence that the source is venous [2, 25]. Because the walls of normal intracranial veins are much thinner than arterial walls, intracranial venous VW-MRI may not be feasible. However, pathological alteration of a vein could make it thicker and visible, and successful wall imaging of larger veins in the leg has been reported [26]. Identifying the source of hemorrhage may also be less important when the source is venous as the prognosis is more likely favorable.
Several studies have assessed the utility of conventional brain MRI in patients with an angiogram-negative nonperimesencephalic SAH, and the diagnostic yield has generally been very low [11, 27]. However, VW-MRI differs considerably from conventional MRI so the results of these studies do not necessarily apply to VW-MRI. The VW-MRI technique we have used has a voxel size of 0.4 × 0.4 × 2.0 mm, and three-dimensional VW-MRI pulse sequences can achieve voxel size of 0.4 × 0.4 × 0.4 mm [28]. This compares with standard brain MRI pulse sequence which typically have voxel size in the order of 0.8 × 0.8 × 5.0 mm. Likely as important as high spatial resolution, VW-MRI employs black-blood pulse sequences which optimize visualization of the arterial wall relative to luminal blood and cerebrospinal fluid. In this study, we have use two-dimensional VW-MRI pulse sequences in multiple planes, and we were able to obtain high quality vessel wall images in short- and long-axis through arteries of interest. However, repeating sequences in multiple planes is time consuming and is not ideal for the assessment of arteries that curve. The relatively long scan duration and the necessity for patient compliance to avoid motion artifact can be problematic in patients with recent SAH, and may prevent sicker patients from undergoing VW-MRI unless they are intubated. Three-dimensional VW-MRI sequences can partly overcome this limitation, since the scanning time is substantially reduced with this technique [28]. The clinical characteristics of the patients in our study are mainly similar to previous cohorts of patients with an angiogram-negative nonperimesencephalic SAH, with a mildly higher incidence of hydrocephalus in our study [2, 6, 29].
13
8
The time between SAH and VW-MRI was longer than 3 months for four patients in our study. This was the result of logistic issues, but a long delay is undesirable as the overall goal is to identify the cause of hemorrhage early enough to prevent recurrent hemorrhage. As well, this delay may have reduced our sensitivity for detecting a cause of hemorrhage as some conditions such as an intracranial arterial dissection can spontaneously heal [30]. However, if VWMRI is performed too early, a large volume of blood in the subarachnoid space may confound identification of arterial wall abnormality on VW-MRI. Based on experience in the study, VW-MRI at 14 days after SAH may provide a reasonable balance between clearing of most blood on T1- and T2-weighetd VW-MRI and imaging early enough to potentially treat a lesion before recurrent hemorrhage. Conclusions To summarize, we explored the utility of high-resolution VW-VRI in the patients with non-perimesencephalic, angiogram-negative SAH. The VW-MRI findings did not alter patient management in our cohort. However, 7 of 11 patients had VW-MRI abnormalities, and given a lack of literature on VW-MRI in patients with angiogram-negative SAH, we hope that our description and discussion of our findings will be useful for other clinicians who perform high-resolution VW-MRI in this patient population in a research or clinical setting. Larger studies are likely warranted to further evaluate the diagnostic yield of VW-MRI in patients with nonperimesencephalic, angiogram-negative SAH. Financial Support JMC is supported by the Netherlands Organisation for Scientific Research (NWO), Thrombosis Foundation Holland, Dutch Heart Foundation, and the Remmert Adriaan Laan Fonds. Conflict of Interest On behalf of all authors, the corresponding author states that there is no conflict of interest.
References 1. van Gijn J, Rinkel GJ. Subarachnoid haemorrhage: diagnosis, causes and management. Brain. 2001;124:249–78. 2. Boswell S, Thorell W, Gogela S, Lyden E, Surdell D. Angiogramnegative subarachnoid hemorrhage: outcomes data and review of the literature. J Stroke Cerebrovasc Dis. 2013;22:750–7. 3. Aviv RI, Shad A, Tomlinson G, Niemann D, Teddy PJ, Molyneux AJ, Byrne JV. Cervical dural arteriovenous fistulae manifesting as subarachnoid hemorrhage: report of two cases and literature review. AJNR Am J Neuroradiol. 2004;25:854–8. 4. Nicastro N, Schnider A, Leemann B. Anaplastic medullary ependymoma presenting as subarachnoid hemorrhage. Case Rep Neurol Med. 2013;2013:701820. 5. van Gijn J, Kerr RS, Rinkel GJ. Subarachnoid haemorrhage. Lancet. 2007;369:306–18.
13
J. M. Coutinho et al. 6. Konczalla J, Schuss P, Platz J, Vatter H, Seifert V, Guresir E. Clinical outcome and prognostic factors of patients with angiogramnegative and non-perimesencephalic subarachnoid hemorrhage: benign prognosis like perimesencephalic SAH or same risk as aneurysmal SAH? Neurosurg Rev. 2015;38:121–7. 7. Maslehaty H, Barth H, Petridis AK, Doukas A, Maximilian Mehdorn H. Special features of subarachnoid hemorrhage of unknown origin: a review of a series of 179 cases. Neurol Res. 2012;34:91–7. 8. Elhadi AM, Zabramski JM, Almefty KK, Mendes GA, Nakaji P, McDougall CG, Albuquerque FC, Preul MC, Spetzler RF. Spontaneous subarachnoid hemorrhage of unknown origin: hospital course and long-term clinical and angiographic follow-up. J Neurosurg. 2014:1–8. 9. Bakker NA, Groen RJ, Foumani M, Uyttenboogaart M, Eshghi OS, Metzemaekers JD, Lammers N, Luijckx GJ, Van Dijk JM. Repeat digital subtraction angiography after a negative baseline assessment in nonperimesencephalic subarachnoid hemorrhage: a pooled data meta-analysis. J Neurosurg. 2014;120:99–103. 10. Agid R, Andersson T, Almqvist H, Willinsky RA, Lee SK, terBrugge KG, Farb RI, Söderman M. Negative CT angiography findings in patients with spontaneous subarachnoid hemorrhage: when is digital subtraction angiography still needed? AJNR Am J Neuroradiol. 2010;31:696–705. 11. Woodfield J, Rane N, Cudlip S, Byrne JV. Value of delayed MRI in angiogram-negative subarachnoid haemorrhage. Clin Radiol. 2014;69:350–6. 12. Andaluz N, Zuccarello M. Yield of further diagnostic work-up of cryptogenic subarachnoid hemorrhage based on bleeding patterns on computed tomographic scans. Neurosurgery. 2008;62:1040–6. 13. Rogg JM, Smeaton S, Doberstein C, Goldstein JH, Tung GA, Haas RA. Assessment of the value of MR imaging for examining patients with angiographically negative subarachnoid hemorrhage. AJR Am J Roentgenol. 1999;172:201–6. 14. Topcuoglu MA, Ogilvy CS, Carter BS, Buonanno FS, Koroshetz WJ, Singhal AB. Subarachnoid hemorrhage without evident cause on initial angiography studies: diagnostic yield of subsequent angiography and other neuroimaging tests. J Neurosurg. 2003;98:1235–40. 15. Matouk CC, Mandell DM, Gunel M, Bulsara KR, Malhotra A, Hebert R, Johnson MH, Mikulis DJ, Minja FJ. Vessel wall magnetic resonance imaging identifies the site of rupture in patients with multiple intracranial aneurysms: proof of principle. Neurosurgery. 2013;72:492–6. 16. Mandell DM, Matouk CC, Farb RI, Krings T, Agid R, terBrugge K, Willinsky RA, Swartz RH, Silver FL, Mikulis DJ. Vessel wall MRI to differentiate between reversible cerebral vasoconstriction syndrome and central nervous system vasculitis: preliminary results. Stroke. 2012;43:860–2. 17. Power S, Matouk C, Casaubon LK, Silver FL, Krings T, Mikulis DJ, Mandell DM. Vessel wall magnetic resonance imaging in acute ischemic stroke: effects of embolism and mechanical thrombectomy on the arterial wall. Stroke. 2014;45:2330–4. 18. Obusez EC, Hui F, Hajj-Ali RA, Cerejo R, Calabrese LH, Hammad T, Jones SE. High-resolution MRI vessel wall imaging: spatial and temporal patterns of reversible cerebral vasoconstriction syndrome and central nervous system vasculitis. AJNR Am J Neuroradiol. 2014;35:1527–32. 19. Rinkel GJ, Wijdicks EF, Hasan D, Kienstra GE, Franke CL, Hageman LM, Vermeulen M, van Gijn J. Outcome in patients with subarachnoid haemorrhage and negative angiography according to pattern of haemorrhage on computed tomography. Lancet. 1991;338:964–8. 20. Bradley WG Jr. MR appearance of hemorrhage in the brain. Radiology. 1993;189:15–26.
High-Resolution Vessel Wall Magnetic Resonance Imaging in Angiogram-Negative 21. Horie N, Morikawa M, Fukuda S, Hayashi K, Suyama K, Nagata I. Detection of blood blister-like aneurysm and intramural hematoma with high-resolution magnetic resonance imaging. J Neurosurg. 2011;115:1206–9. 22. Gonzalez AM, Narata AP, Yilmaz H, Bijlenga P, Radovanovic I, Schaller K, Lovblad KO, Pereira VM. Blood blister-like aneurysms: single center experience and systematic literature review. Eur J Radiol. 2014;83:197–205. 23. Aoki S, Shirouzu I, Sasaki Y, Okubo T, Hayashi N, Machida T, Hoshi E, Suzuki K, Funada N, Araki T. Enhancement of the intracranial arterial wall at MR imaging: relationship to cerebral atherosclerosis. Radiology. 1995;194:477–81. 24. Mossa-Basha M, Hwang WD, De Havenon A, Hippe D, Balu N, Becker KJ, Tirschwell DT, Hatsukami T, Anzai Y, Yuan C. Multicontrast high-resolution vessel wall magnetic resonance imaging and its value in differentiating intracranial vasculopathic processes. Stroke. 2015;46:1567–73. 25. van der Schaaf IC, Velthuis BK, Gouw A, Rinkel GJ. Venous drainage in perimesencephalic hemorrhage. Stroke. 2004;35:1614–8. 26. Ambrose N, Pierce IT, Gatehouse PD, Haskard DO, Firmin DN. Magnetic resonance imaging of vein wall thickness in patients with Behcet’s syndrome. Clin Exp Rheumatol. 2014;32:S99–102.
9
27. Maslehaty H, Petridis AK, Barth H, Mehdorn HM. Diagnostic value of magnetic resonance imaging in perimesencephalic and nonperimesencephalic subarachnoid hemorrhage of unknown origin. J Neurosurg. 2011;114:1003–7. 28. Qiao Y, Steinman DA, Qin Q, Etesami M, Schar M, Astor BC, Wasserman BA. Intracranial arterial wall imaging using threedimensional high isotropic resolution black blood MRI at 3.0 T. J Magn Reson Imaging. 2011;34:22–30. 29. Konczalla J, Platz J, Schuss P, Vatter H, Seifert V, Guresir E. Non-aneurysmal non-traumatic subarachnoid hemorrhage: patient characteristics, clinical outcome and prognostic factors based on a single-center experience in 125 patients. BMC Neurol. 2014;14:140. 30. Wang Y, Lou X, Li Y, Sui B, Sun S, Li C, Jiang P, Siddiqui A, Yang X. Imaging investigation of intracranial arterial dissecting aneurysms by using 3 T high-resolution MRI and DSA: from the interventional neuroradiologists’ view. Acta Neurochir (Wien). 2014;156:515–25.
13
