You are here

Article Text

Article menu

Download PDFPDF

Research paper
Cerebrovascular reactivity within perfusion territories in patients with an internal carotid artery occlusion
  1. Reinoud P H Bokkers1,
  2. Matthias J P van Osch2,
  3. Catharina J M Klijn3,
  4. L Jaap Kappelle3,
  5. Jeroen Hendrikse1
  1. 1Department of Radiology, University Medical Center Utrecht, Utrecht, The Netherlands
  2. 2Department of Radiology, CJ Gorter Center for High-Field MRI, Leiden University Medical Center, The Netherlands
  3. 3Department of Neurology, Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht, Utrecht, The Netherlands
  1. Correspondence to Dr R P H Bokkers, Department of Radiology, University Medical Center Utrecht, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands; r.p.h.bokkers{at}umcutrecht.nl

Abstract

Background Arterial spin labelling (ASL) is an MRI technique for measuring perfusion at the brain tissue level. The aim of the study was to investigate cerebrovascular reactivity (CVR) at brain-tissue level in patients with an internal carotid artery (ICA) occlusion by combining ASL-MRI with a vascular challenge, and determine whether the CVR varies within the perfusion territory of the brain-feeding arteries.

Methods Sixteen patients with a symptomatic ICA occlusion and 16 age-matched healthy control subjects underwent perfusion and perfusion-territory selective ASL-MRI before and after acetazolamide administration. CVR was assessed throughout the brain in the grey matter supplied by the unaffected asymptomatic ICA and the basilar artery.

Results Cerebral blood flow increased (p<0.01) in all perfusion territories after acetazolamide in the patients and controls. In the tissue supplied by the unaffected contralateral ICA, CVR was lower in the tissue supplied by the unaffected contralateral ICA in the patients when compared with the controls (22.8±16.1 vs 54.2±13.1%; mean difference, −31.5%, 95% CI −42.1 to −20.8). Within the perfusion territory of the unaffected ICA, the CVR was lower in the brain tissue on the side of the occluded ICA than on the side of the unaffected ICA (13.5±20.4 vs 26.2±16.0%; paired mean difference −12.5%, 95% CI −20.3 to −4.7).

Conclusion ASL-MRI can assess impaired cerebrovascular reactivity at the brain-tissue level in patients with a symptomatic ICA occlusion. Assessment of CVR with ASL-MRI may help to identify the tissue most at risk for future stroke and as such may guide medical treatment.

  • Cerebrovascular disease
  • cerebral blood flow
  • MRI

https://doi.org/10.1136/jnnp.2010.233338

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Patients with a symptomatic internal carotid artery (ICA) occlusion have an annual risk of 5–6% for recurring stroke.1 This risk is raised to 9–18% per year in patients with compromised cerebral haemodynamics and poor collateral blood flow.2–4 A large, international randomised trial in 1985 showed that extracranial-to-intracranial (EC/IC) bypass surgery does not prevent stroke in patients with symptomatic ICA occlusion.5 Additional studies have however suggested that patients with haemodynamically compromised brain perfusion may benefit from bypass surgery.6

    Haemodynamic compromise occurs when the compensatory responses of the brain to a decrease in the perfusion pressure are exhausted, and adequate cerebral blood flow (CBF) cannot be maintained.7 With a steno-occlusive lesion in one or more brain-feeding arteries, additional flow is initially recruited through collateral pathways.8 When this flow is not sufficient, resistance arterioles dilate in order to reduce the vascular resistance to arterial inflow. The vasodilatory capacity, or cerebrovascular reactivity, can be assessed indirectly by measuring the dilatory response after a vascular challenge. This can be measured either at the brain-tissue level with techniques, such as positron emission tomography (PET) and single-photon emission tomography, or by measuring the increase in flow velocity in the middle cerebral artery with transcranial Doppler (TCD).2 3 9 Both approaches have limitations. Measurements at the brain-tissue level are non-specific to the route of blood supply, and flow-velocity measurements are limited to single brain-feeding arteries. In patients with a steno-occlusive lesion in one of the ICAs, this is especially problematic, as additional blood is recruited through collateral pathways, and there is a shift in the perfusion territories of the brain-feeding arteries.10 11

    Arterial spin-labelling perfusion MRI (ASL-MRI), in combination with a vascular challenge, has been introduced as an alternative technique for measuring cerebrovascular reactivity at the brain-tissue level.12 13 ASL-MRI uses radiofrequency pulses to magnetically label blood and does not require gadolinium-based contrast agents. With the recent introduction of perfusion-territory selective labelling techniques, ASL-MRI can also visualise the individual contribution of the cerebral arteries and collateral vessels to the brain.14 15

    The objective of this study was to assess cerebrovascular reactivity at the brain-tissue level with ASL-MRI in patients with an ICA occlusion and in healthy controls, and determine whether cerebrovascular reactivity varies within the perfusion territories of the brain-feeding arteries of patients with an ICA occlusion.

    Methods

    The study was conducted with institutional ethical standards committee approval, and all participants gave written informed consent according to the Declaration of Helsinki.

    Subjects

    Sixteen patients (mean age 56.3 years±13.8 (SD)) with a recently symptomatic atherosclerotic ICA occlusion and 16 age-matched healthy control volunteers (mean age 56.5±5.7) were prospectively included into the study. All patients were admitted to our hospital because of a transient ischaemic attack (TIA) or non-disabling stroke on the side of the ICA occlusion. Imaging was performed within 3 months of symptom onset. Patients were excluded from the study if they had diabetes mellitus or severe renal or liver dysfunction, or had experienced a stroke causing a major disability (modified Rankin score of 3–5) in the past.16 The ICA occlusion was diagnosed with either CT or MR angiography. The healthy control volunteers were recruited through local media advertisements and did not have any history of neurological disease or vascular pathology on MRI or MR angiography.

    MRI

    All examinations were performed on a clinical 3T MRI scanner (Achieva, Philips Medical Systems, Best, The Netherlands). Perfusion images were obtained before and 15 min after administration of a bolus of 14 mg/kg, with a maximum dose of 1200 mg, acetazolamide (Goldshield Pharmaceuticals, Croydon, UK). Perfusion imaging was performed with a pseudo-continuous ASL sequence in combination with background suppression.17 For positioning of the imaging section, a low-resolution T1-weighted spin-echo sequence was obtained in the sagittal plane. The perfusion images consisted of seventeen 7 mm slices aligned parallel to the orbitomeatal angle, acquired in ascending fashion with an in-plane resolution of 3×3 mm (true acquisition resolution). The other ASL-MRI parameters were: repetition time (TR), 4000 ms; echo time (TE), 14 ms; pairs of control/label, 38; postlabelling delay, 1525 ms; field of view (FOV), 240×240×119 mm2; matrix, 80×79; SENSE, 2.5; scan time, 5 min.

    The perfusion territories of the carotid arteries and the basilar artery were assessed with a perfusion-territory selective ASL sequence using pseudo-continuous tagging, according to a previously published protocol.18 Selective ASL labelling (same labelling settings and geometry as the ASL perfusion scan) was accomplished by manipulating the spatial labelling efficiency by applying additional gradients between the labelling pulses. The additional gradients were applied in sets of five dynamics: no labelling applied (control), non-selective labelling applied (globally perfusion weighted), labelling varied in the right–left (RL) direction (distance of 50 mm between full label and control situation), labelling varied in the anterior–posterior (AP1) direction (distance of 18 mm between full label and control situation) and labelling varied in the AP direction (AP2, similar to AP1, but shifted 9 mm in posterior direction compared with the previous dynamic).

    An inversion recovery sequence was acquired prior to the perfusion scans to measure the magnetisationmagnetisation of arterial blood, the M0 which is needed to quantitatively calculate CBF in ml/100 ml/min, and to segment brain tissue into grey and white matter. Both the perfusion-territory selective ASL and inversion recovery sequence were acquired with echo planar imaging with the same geometry and resolution as the ASL images. T2-weighted fluid-attenuated inversion recovery (FLAIR) images and a three-dimensional time-of-flight MR angiography with subsequent maximum-intensity projection reconstruction were acquired with standard imaging sequences provided by the MRI vendor.

    The presence of collateral flow routes in the circle of Willis were assessed with a three-dimensional time-of-flight MR angiography sequence with subsequent maximal intensity projection reconstruction (TR/TE 23/3.5 ms; flip angle 18°; FOV 200×200 mm2; matrix 304×200; 100 slices; slice thickness 1.2 mm with 0.6 mm overlap; scan time, 3 min). The direction of blood flow in these collaterals was determined according to a previously published imaging protocol with two consecutive two-dimensional phase-contrast MR imaging measurements, one of which was phase-encoded in the left–right direction and one in the antero-posterior direction (TR/TE 9.4/5.9 ms; flip angle 7.5°; FOV 250×188 mm; matrix 256×134; eight averages; slice thickness 13 mm; velocity sensitivity 40 cm/s, scan time 20 s).19 Anterior collateral flow was defined as flow across the anterior communicating artery with retrograde flow in the precommunicating part of the anterior cerebral artery (A1 segment). Posterior-to-anterior flow in the posterior communicating artery was considered to represent posterior collateral flow.

    Data analysis

    Data were analysed with Matlab (version 7.5) and SPM5. CBF images (ml/100 ml/min) were calculated from the ASL-MR images according to a previously published model.20 The T2* transversal relaxation rate and T1 of arterial blood at 3T were assumed to be, respectively, 50 ms and 1680 ms.21 22 The blood water content was assumed to be 0.76%.23 The blood magnetisation at thermal equilibrium (M0) for all volunteers was determined by selecting a region of interest in the cerebral spinal fluid and iteratively fitting the inversion recovery data by a non-linear least-squares method.23

    The CBF of the grey matter of the perfusion territories of the basilar and ICA was measured before and after administration of acetazolamide. Cerebrovascular reactivity was defined as the percentage of CBF increase after administration of acetazolamide. For the placement of the regions of interest in the grey matter throughout all 17 slices, three preprocessing steps were performed (figure 1). First, to avoid partial voluming of white matter, a surrogate T1-weighted image was calculated from the inversion recovery sequence by calculating the reciprocal of the quantitative T1. This was segmented into grey- and white-matter probability maps with SPM, and corrective thresholding was subsequently applied to ensure maximal exclusion of all white matter. Second, the perfusion territories of the basilar and ICA were identified by means of a k-means clustering algorithm.24 The individual perfusion territories were then manually drawn on the output images of the clustering algorithm. Two additional regions of interest (ROIs) were drawn within the perfusion territory of the contralateral ICA: one on the unaffected side and one in the affected hemisphere on the side of the occlusion. This ROI represents the brain tissue supplied through collateral pathways originating from the asymptomatic ICA. For this, the perfusion-territory crossing over the midline of the brain into the hemisphere with the ICA occlusion was delineated (figure 1). The final step was to combine the grey-matter masks with the segmented perfusion-territory masks. Areas of hyperintensities on FLAIR, depicting areas of infarction, were manually excluded from the ROI. To correct for motion, all images pre- and postacetazolamide were first coregistered with SPM to the baseline CBF map using the normalised mutual information and a rigid body transformation.

    Figure 1
    Figure 1

    Pictorial description of the preprocessing steps in three of the 17 slices in a patient with an occluded right internal carotid artery (ICA). The upper two rows are cerebral blood flow (CBF) maps (ml/100 ml/min). In the selective arterial spin labelling image, the perfusion territory of the left ICA is portrayed in green and the basilar artery in red. By combining the segmented grey-matter map with the perfusion-territory information, a mask was obtained of the grey matter perfused by the ICA, basilar artery and collateral pathways originating from the unaffected contralateral ICA.

    SPSS (version 15) was used for statistical analysis. Differences between pre- and postacetazolamide CBF measurements, and between the perfusion territories within the healthy control subject, were assessed using a paired t test. Because no differences were found in the CBF or cerebrovascular reactivity between the left and right ICAs in the healthy control subjects, values for both perfusion territories were averaged for further comparisons. To compare the cerebrovascular reactivity measurements in the patients with the healthy control subjects, an independent-samples t test was used. A paired t test was used to compare cerebrovascular reactivity measurements of the perfusion territories in patients. The mean differences in cerebrovascular reactivity with the 95% CIs were calculated and considered significantly different if the 95% CI did not include zero. Values are expressed as mean±SD.

    Results

    The demographic and clinical characteristics of the subjects are outlined in table 1. Figure 2 shows CBF maps pre- and postacetazolamide of a 47-year-old man with a symptomatic occlusion of the left ICA and collateral blood flow from the unaffected contralateral ICA. Figure 3 shows CBF maps of a 51-year-old man with a symptomatic occlusion of the right ICA and collateral blood flow from the posterior circulation. Both patients had decreased CBF and cerebrovascular reactivity on the side of the occluded ICA.

    View this table:
    Table 1

    Demographic and clinical characteristics of the study population

    Figure 2
    Figure 2

    Perfusion images of a 47-year-old man with a symptomatic occlusion of the left carotid artery and collateral flow from unaffected contralateral internal carotid artery (green). Cerebral blood flow (CBF) and cerebrovascular reactivity are decreased in the left hemisphere. On the anatomical fluid-attenuated inversion recovery (FLAIR) image, multiple hyperintensities are present in the left hemisphere. MRA, MR angiography.

    Figure 3
    Figure 3

    Perfusion images of a 51-year old man with a symptomatic occlusion of the right carotid artery and collateral flow from the posterior circulation (red). Cerebral blood flow (CBF) and cerebrovascular reactivity are decreased in the right hemisphere. On the anatomical fluid-attenuated inversion recovery (FLAIR) image, multiple hyperintensities are present in the right hemisphere. MRA, MR angiography.

    Table 2 summarises the CBF measurements before and after acetazolamide, and the cerebrovascular reactivity measurements. CBF increased (p<0.01) in all perfusion territories after administration of acetazolamide. In the patients with an ICA occlusion, the cerebrovascular reactivity was lower in the tissue fed by the unaffected ICA (22.8±16.1%) when compared with the healthy control subjects (54.2±13.1%; mean difference −31.5%; 95% CI −42.1 to −20.8). The cerebrovascular reactivity was also lower in the brain tissue fed by the basilar artery (32.2±16.6%) when compared with the healthy control subjects (73.7±23.7%; mean difference −40%; 95% CI −55 to −25).

    View this table:
    Table 2

    Cerebral blood flow and cerebrovascular reactivity in the carotid and basilar artery

    When compared with the healthy control subjects, in patients the cerebrovascular reactivity was lower within the perfusion territory of the unaffected ICA; both in the hemisphere ipsilateral to the ICA occlusion (13.5±20.4 versus 54.2±13.1%; mean difference −40.7%, 95% CI −53.2 to −28.2) and in the hemisphere ipsilateral to the unaffected side (26.2±16.0 versus 54.2±13.1%; mean difference −28.0%, 95% CI −38.6 to −17.5). Within the perfusion territory, the cerebrovascular reactivity was lower in the brain tissue on the side of the occlusion (13.5±20.4%) than on the side of the unaffected ICA (26.2±16.0%; paired mean difference −12.5%, 95% CI −20.3 to −4.7). Nine of the 16 patients had severely decreased cerebrovascular reactivity (<20%) in the hemisphere ipsilateral to the ICA occlusion (3.7±6.6 vs 28.3±25.6%; mean difference 24.6%; 95% CI 5.5 to 43.7). In four of the 16 patients, CBF decreased in the brain tissue on the side of the occlusion after acetazolamide (−6.0± 9.9 vs 20.6±18.6%; mean difference 26.7%, 95% CI 5.3 to 48.0).

    Time-of-flight and phase-contrast angiography demonstrated anterior collateral flow in three patients, posterior collateral flow in six patients and a combination of both in four patients. In three patients, it could not be defined, owing to incorrect planning of the imaging section. The presence or absence of anterior or posterior collateral flow did not affect CBF or the vascular reactivity in the hemisphere with the ICA occlusion.

    Discussion

    We investigated the cerebrovascular reactivity of the brain-feeding arteries at the level of the brain tissue by combining quantitative and perfusion-territory selective ASL-MRI with a vascular challenge. Our study shows that the cerebrovascular reactivity is decreased throughout the brain in patients with a symptomatic ICA occlusion and varies within the perfusion territory of the unaffected contralateral ICA. Cerebrovascular reactivity on the side of the ICA occlusion fed by the unaffected contralateral ICA is the most impaired.

    Previous studies have shown that the presence of primary and secondary collaterals is associated with decreased cerebrovascular reactivity.25 26 Primary collaterals are the anterior and posterior communicating arteries of the circle of Willis, and the ophthalmic artery and leptomeningeal vessels are considered to be secondary collaterals.11 A fully developed collateral network, however, has also been reported to be associated with normal cerebrovascular reactivity and a lower risk of future stroke when compared with patients without collateral or with only primary collaterals.27 This discrepancy could possibly be explained by the techniques used to evaluate the contribution of collaterals. The efficiency of the collateral vasculature cannot be evaluated by merely determining the presence of certain collateral pathways, but should ideally be evaluated by measuring the haemodynamic status in tissue fed by these collaterals as well. Our results show that the cerebrovascular reactivity is the most impaired on the side of the ICA occlusion in tissue that is fed by collaterals originating from the unaffected contralateral ICA.

    Although there are no previous studies validating ASL-MRI perfusion reactivity measurements by comparing with other established techniques, our finding of impaired cerebrovascular reactivity in patients with a symptomatic ICA occlusion is in agreement with other studies that have used either TCD to measure cerebrovascular reactivity in the intracranial arteries or use techniques such as PET and SPECT to measure cerebrovascular reactivity at the brain tissue level.28 29 ASL-MRI has previously been used to assess cerebrovascular reactivity in patients with large artery cerebrovascular stenosis of the anterior circulation.12 With a continuous ASL-MRI sequence at 1.5 T in combination with acetazolamide, a varying pattern of cerebrovascular reactivity was observed, ranging from normal to focal and diffuse haemodynamic impairment. In a test–retest reproducibility study, Yen et al have furthermore shown that cerebrovascular reactivity measurements performed with a flow sensitive alternating inversion recovery ASL technique at 1.5 T has a good reproducibility and is sensitive to small changes related to disease or treatment.13 The advantage of the currently used pseudo-continuous ASL sequence at 3 T is the higher signal-to-noise ratio when used in combination with background suppression. For cerebrovascular reactivity measurements, this is of particular importance, as a relatively small increase in signal has to be measured. Furthermore, by adding perfusion-territory selective ASL, we were able to assess the perfusion territories of the cerebral arteries. We were therefore able to assess cerebrovascular reactivity within the perfusion territories of the brain-feeding arteries and evaluate the tissue fed through collaterals. The slightly lower cerebrovascular reactivity measured in the hemisphere contralateral to the ICA occlusion when compared with the healthy control subjects might potentially be explained by the presence of a contralateral ICA stenosis in some of the patients or decreased cerebrovascular reactivity owing to generalised atherosclerosis.

    For patients with a symptomatic occlusion of the ICA, the best management is yet to be defined.30 EC/IC bypass surgery has been largely abandoned, since the EC-IC study has shown that it does not prevent stroke.31 It has however been suggested that EC/IC bypass surgery may be effective in a subgroup of patients with severely impaired cerebral haemodynamics.32 33 In the Carotid Occlusion Surgery Study efficacy of the EC/IC bypass is currently being studied in patients with increased oxygen extraction fraction as measured by PET.34 A drawback of PET, however, is that it is only available in a limited number of institutions. ASL-MRI can simultaneously assess the cerebrovascular reactivity and the shifts in the perfusion territories of the brain-feeding arteries at tissue level on MRI scanners widely available in a clinical setting. It may therefore help depict the brain tissue most at risk for future stroke and define which patients would benefit most from revascularisation operations such as EC/IC bypass or carotid endarterectomy of a severe contralateral ICA stenosis. With further studies, this could be either a certain cerebrovascular reactivity threshold or the occurrence of steal, where blood is redistributed to more healthy tissue and CBF decreases.

    A limitation of this study is that in patients with an ICA occlusion, collateralisation may lead to a delayed arrival of the bolus of magnetically labelled blood spins. As the perfusion-weighted images are acquired after a fixed amount of time in ASL-MRI, this could potentially lead to an underestimation of the perfusion. And in healthy tissue with uncompromised cerebrovascular reactivity, the blood may flow faster from the supplying arteries to the brain tissue, resulting in complete arrival of labelled spins during imaging and thus a stronger ASL signal when compared with impaired tissue. During the acetazolamide challenge, these transit times might decrease owing to faster flow, leading to a relative less severe underestimation of ASL signal and an overestimation of cerebrovascular reactivity. To minimise these effects, we employed a delay time of 1525 ms in this study, where the protons that have been labelled directly after the start of the labelling have an effective delay time of more than 3 s. Furthermore, these transit time effects would lead to an overestimation of the cerebrovascular reactivity ipsilateral to the occlusion, whereas in this study we observed a statistically decreased reactivity in the patients. Collateral flow from the posterior circulation was not evaluated in this study, as the brain tissue that is supplied through collaterals originating from the vertebrobasilar arteries could not be reliably differentiated from normal variations in the perfusion territories of the posterior circulation. Perfusion-territory selective ASL-MRI can visualise the extent of the perfusion territory of the vertebrobasilar artery; however, the boundary between tissue natively being perfused by the basilar artery and tissue supplied through collaterals (posterior communicating artery or leptomeningeals) cannot be accurately determined. Finally, two patients were included with a contralateral stenosis of ≥70. Although this may have decreased the cerebrovascular reactivity measurements in the unaffected contralateral ICA, we still found that the cerebrovascular reactivity varies within the perfusion territory of the unaffected artery.

    ASL-MRI can assess impaired cerebrovascular reactivity at the brain tissue level in patients with a symptomatic ICA occlusion. Impairment of the haemodynamic status of the brain is associated with increased risk of stroke. With further research, cerebrovascular reactivity assessment with ASL-MRI may possibly help identify the brain tissue most at risk for future stroke and as such may guide medical treatment.

    References

    1. ↵
      1. Klijn CJ,
      2. Kappelle LJ,
      3. Algra A,
      4. et al
      . Outcome in patients with symptomatic occlusion of the internal carotid artery or intracranial arterial lesions: a meta-analysis of the role of baseline characteristics and type of antithrombotic treatment. Cerebrovasc Dis 2001;12:228–34.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Vernieri F,
      2. Pasqualetti P,
      3. Passarelli F,
      4. et al
      . Outcome of carotid artery occlusion is predicted by cerebrovascular reactivity. Stroke 1999;30:593–8.
      OpenUrlAbstract/FREE Full Text
    3. ↵
      1. Grubb RL Jr.,
      2. Derdeyn CP,
      3. Fritsch SM,
      4. et al
      . Importance of hemodynamic factors in the prognosis of symptomatic carotid occlusion. JAMA 1998;280:1055–60.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Klijn CJ,
      2. Kappelle LJ,
      3. van Huffelen AC,
      4. et al
      . Recurrent ischemia in symptomatic carotid occlusion: prognostic value of hemodynamic factors. Neurology 2000;55:1806–12.
      OpenUrlAbstract/FREE Full Text
    5. ↵
      The EC/IC bypass study group. Failure of extracranial–intracranial arterial bypass to reduce the risk of ischemic stroke. Results of an international randomized trial. N Engl J Med 1985;313:1191–200.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Adams HP Jr.
      . Occlusion of the internal carotid artery: reopening a closed door? JAMA 1998;280:1093–4.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Derdeyn CP,
      2. Videen TO,
      3. Yundt KD,
      4. et al
      . Variability of cerebral blood volume and oxygen extraction: stages of cerebral haemodynamic impairment revisited. Brain 2002;125:595–607.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Powers WJ
      . Cerebral hemodynamics in ischemic cerebrovascular disease. Ann Neurol 1991;29:231–40.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Hirano T,
      2. Minematsu K,
      3. Hasegawa Y,
      4. et al
      . Acetazolamide reactivity on 123I-IMP single photon emission computed tomography in patients with major cerebral artery occlusive disease: correlation with positron emission tomography parameters. J Cereb Blood Flow Metab 1994;14:763–70.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. van Laar PJ,
      2. Hendrikse J,
      3. Klijn CJ,
      4. et al
      . Symptomatic carotid artery occlusion: flow territories of major brain-feeding arteries. Radiology 2007;242:526–34.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Liebeskind DS
      . Collateral circulation. Stroke 2003;34:2279–84.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Detre JA,
      2. Samuels OB,
      3. Alsop DC,
      4. et al
      . Noninvasive magnetic resonance imaging evaluation of cerebral blood flow with acetazolamide challenge in patients with cerebrovascular stenosis. J Magn Reson Imaging 1999;10:870–5.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Yen YF,
      2. Field AS,
      3. Martin EM,
      4. et al
      . Test–retest reproducibility of quantitative CBF measurements using FAIR perfusion MRI and acetazolamide challenge. Magn Reson Med 2002;47:921–8.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Paiva FF,
      2. Tannus A,
      3. Silva AC
      . Measurement of cerebral perfusion territories using arterial spin labelling. NMR Biomed 2007;20:633–42.
      OpenUrlCrossRefPubMed
    15. ↵
      1. van Laar PJ,
      2. van der Grond J,
      3. Hendrikse J
      . Brain perfusion territory imaging: methods and clinical applications of selective arterial spin-labeling MR imaging. Radiology 2008;246:354–64.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Bamford JM,
      2. Sandercock PAG,
      3. Warlow CP,
      4. et al
      . Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1989;20:828.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Dai W,
      2. Garcia D,
      3. de Bazelair C,
      4. et al
      . Continuous flow-driven inversion for arterial spin labeling using pulsed radio frequency and gradient fields. Magn Reson Med 2008;60:1488–97.
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Gevers S,
      2. Nederveen A,
      3. Bokkers RP,
      4. et al
      . Reproducibility of flow territories defined by plannings-free vessel encoded pseudo-continuous arterial spin labeling. Proceedings 17th Scientific Meeting, International Society for Magnetic Resonance in Medicine 2009;1524.
    19. ↵
      1. Rutgers DR,
      2. Klijn CJ,
      3. Kappelle LJ,
      4. et al
      . A longitudinal study of collateral flow patterns in the circle of Willis and the ophthalmic artery in patients with a symptomatic internal carotid artery occlusion. Stroke 2000;31:1913–20.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Alsop DC,
      2. Detre JA
      . Reduced transit-time sensitivity in noninvasive magnetic resonance imaging of human cerebral blood flow. J Cereb Blood Flow Metab 1996;16:1236–49.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Golay X,
      2. Petersen ET,
      3. Hui F
      . Pulsed star labeling of arterial regions (PULSAR): a robust regional perfusion technique for high field imaging. Magn Reson Med 2005;53:15–21.
      OpenUrlCrossRefPubMed
    22. ↵
      1. St Lawrence KS,
      2. Wang J
      . Effects of the apparent transverse relaxation time on cerebral blood flow measurements obtained by arterial spin labeling. Magn Reson Med 2005;53:425–33.
      OpenUrlCrossRefPubMed
    23. ↵
      1. Chalela JA,
      2. Alsop DC,
      3. Gonzalez-Atavales JB,
      4. et al
      . Magnetic resonance perfusion imaging in acute ischemic stroke using continuous arterial spin labeling. Stroke 2000;31:680–7.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Wong EC,
      2. Kansagra AP
      . Mapping middle cerebral artery branch territories with vessel encoded pseudo-continuous ASL: sine/cosine tag modulation and data clustering in tagging efficiency space. Proceedings 16th Scientific Meeting, International Society for Magnetic Resonance in Medicine 2009.
    25. ↵
      1. Norrving B,
      2. Nilsson B,
      3. Risberg J
      . rCBF in patients with carotid occlusion. Resting and hypercapnic flow related to collateral pattern. Stroke 1982;13:155–62.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Muller M,
      2. Schimrigk K
      . Vasomotor reactivity and pattern of collateral blood flow in severe occlusive carotid artery disease. Stroke 1996;27:296–9.
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Vernieri F,
      2. Pasqualetti P,
      3. Matteis M,
      4. et al
      . Effect of collateral blood flow and cerebral vasomotor reactivity on the outcome of carotid artery occlusion. Stroke 2001;32:1552–8.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Imaizumi M,
      2. Kitagawa K,
      3. Hashikawa K,
      4. et al
      . Detection of misery perfusion with split-dose 123I-iodoamphetamine single-photon emission computed tomography in patients with carotid occlusive diseases. Stroke 2002;33:2217–23.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Ogasawara K,
      2. Ogawa A,
      3. Yoshimoto T
      . Cerebrovascular reactivity to acetazolamide and outcome in patients with symptomatic internal carotid or middle cerebral artery occlusion: a xenon-133 single-photon emission computed tomography study. Stroke 2002;33:1857–62.
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Flemming KD,
      2. Brown RD Jr.,
      3. Petty GW,
      4. et al
      . Evaluation and management of transient ischemic attack and minor cerebral infarction. Mayo Clin Proc 2004;79:1071–86.
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      1. Caplan LR,
      2. Piepgras DG,
      3. Quest DO,
      4. et al
      . EC–IC bypass 10 years later: is it valuable? Surg Neurol 1996;46:416–23.
      OpenUrlPubMedWeb of Science
    32. ↵
      1. Adams HP Jr.
      . Occlusion of the internal carotid artery: reopening a closed door? JAMA 1998;280:1093–4.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Derdeyn CP,
      2. Grubb RL,
      3. Powers WJ
      . Cerebral hemodynamic impairment: methods of measurement and association with stroke risk. Neurology 1999;53:251–9.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Grubb RL Jr.,
      2. Powers WJ,
      3. Derdeyn CP,
      4. et al
      . The carotid occlusion surgery study. Neurosurg Focus 2003;14:e9.
      OpenUrlPubMed

    Footnotes

    • Funding MJPvO receives support from the Technology Foundation STW, applied science division of NWO and the technology programme of the Ministry of Economic Affairs. JH receives support from The Netherlands Organization for Scientific Research (no 916-76-035).

    • Competing interests None.

    • Patient consent Obtained.

    • Ethics approval Ethics approval was provided by the Institutional Ethical Standards Committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.