Surgical and Radiologic Anatomy

, Volume 27, Issue 6, pp 544–547 | Cite as

The value of phase-contrast magnetic resonance angiography of the circle of Willis in predicting cerebral ischemia-hypoxia (shunt need) during carotid endarterectomy

Radiologic Anatomy


Background: Cerebral perfusion during carotid cross-clamping depends on collateral function of the circle of Willis. The aim of this study was to determine the value of 3D Phase-Contrast (3D PC) MR angiography in predicting pre-operatively the need of shunting. Methods: 3D PC MR angiography were performed in 121 patients before carotid endarterectomy under locoregional anaesthesia. Based on the MR analysis, the risk of cerebral ischemia–hypoxia during clamping was classified as high, moderate and low. The analysis was then correlated with intraoperative neurological examination. Results: In patients with high risk of cerebral ischemia (n=9), immediate cerebral ischemia developed in all patients within one min of clamping (P<0.001). All nine underwent shunt placement. In six of the patients with moderate risk (n=46), cerebral ischemia occurred between 20 and 25 min after clamping; All occurred during intraoperative hypotension. Conclusion: 3D PC MR angiography can significantly determine the need of shunting in patients with important risk of immediate intraoperative cerebral ischemia. It also focuses on the intraoperative blood pressure stability in patients with moderate risk of ischemia.


Cerebral ischemia Magnetic Resonance angiography Carotid surgery Carotid artery Circle of Willis Anatomy 


The clinical tolerance of acute carotid artery cross-clamping during carotid endarterectomy (CE) under locoregional anaesthesia depends on the functionality of the intracranial cerebral collateral circulation and on the blood pressure stability [16]. The collateral flow to the cerebral hemisphere is mainly through the circle of Willis. Lazorthes and Gouaze [9] have described 22 different configurations of the circle of Willis. The typical normal polygon configuration is present only in 50% of the general population [1, 3, 9, 12]. Moreover, the collateral circulation is not functionnal and require placement of a shunt in about 10% of patients undergoing CE [5]. The use of an intraluminal shunt can be performed systematically or selectively. It appears that shunt placement complications such as dissection or thromboembolism might be reduced by selective-shunting only for severe persisting ischemia during carotid clamping [6]. The traditional methods used to assess pre-operatively the need for shunt placement are conventional angiography and transcranial Doppler ultrasonography [2, 6, 7, 19, 20, 21]. The purpose of this study was to determine the value of 3D phase contrast magnetic resonance angiography (3D PC MR) in predicting pre-operatively the risk of severe cerebral ischemia and therefore the need of shunting.

Material and methods

Patient population

From October 1998 to December 2001, 121 patients benefited from pre-operative 3D PC intracranial MR angiograms. There were 95 male and 26 female patients with a mean age of 69 years (range 49–89 years). The criteria of the North American Symptomatic Carotid Endarterectomy Trial were used for the determination of percent stenosis. There were 45 symptomatic patients with >70% carotid stenosis on duplex scanning and 76 asymptomatic patients with >80% carotid stenosis on duplex scanning. Seven patients had a contralateral internal carotid occlusion.

Pre-operative evaluation

Pre-operative MR angiography was performed with MR imaging system. Radiologist evaluated the collateral flow and morphology on pre-operative 3D PC MR angiograms in the anterior part and posterior part of the circle of Willis (Fig. 1).
Fig. 1

Typical normal polygon configuration of the circle of Willis. M1 indicates main trunk of the middle cerebral artery;A1, pre-communicating part, A2, post-communicating part of the anterior cerebral artery; P1, precommunicating part, P2, post-communicating part of the posterior cerebral artery; Com P: Posterior communicating artery; *: anterior communicating artery; **: ophthalmic artery

The risk of cerebral ischemia was considered as high when flow was absent or weak in a major segment of both anterior (contralateral M1, A1 controlateral, Com Ant) and posterior (Com Post, P1) collateral pathway of cerebral blood flow. The risk was considered as moderate when flow was absent or weak in a segment of anterior or posterior collateral pathway and it was considered as low when flow was normal in the collateral circulation.

Anaesthetic and surgical techniques

Baseline blood pressure was recorded during anaesthetic examination 1 week before operation. Patients underwent a superficial and deep cervical plexus block especially during carotid clamping. Neurologic evaluation and blood pressure control were repeated during carotid clamping. Intraluminal shunts were placed whenever neurologic deterioration occurred.

Statistical analysis

The risk of cerebral ischemia was correlated with the history of severe ischemia requiring shunt placement during carotid clamping and with the blood pressure level. Statistical analysis were performed using computerized software (Statwiew, brain power Inc, Calabass, CA, USA). Statistical significance was evaluated with the χ2 test. P values less than 0.05 were considered significant.


Table 1 shows the distribution of anomalies observed on Angio-MRI compared with anomalies described in the literature. The frequency of complete circle of Willis and of anomalies was similar to that reported in anatomical studies. Relationship between predictive risk of shunt need and frequency of cerebral ischemia during carotid clamping is depicted in Table 2. Statistical analysis revealed that the number of immediate intraoperative ischemia was significantly increased in patient with high-risk of shunt need defined on MRI (P<0.001). Cerebral ischemia developed in all patients within 1 min of clamping despite a high blood pressure level. Polygon configuration at risk of immediate ischemia were type 5 (n=2), type 14 (n=2), type 15 (n=1) according to Lazorthe’s classification [9]. Four of the seven patients with controlateral carotid occlusion experienced immediate ischemia. These four patients presented an incompleteness of the posterior part of the circle of Willis. Delayed ischemia occurred during intraoperative hypotension in six patients with moderate risk of ischemia.
Table 1

Comparison between anomalies observed on 3D PC MRI and reported on anatomical studies (Polygons configurations according to Lazorthes classification [9])

Polygon configuration

Arterial segment hypoplastic or absent

Observed on Angio-MRI % (n patients)

In anatomical studies [1, 8, 11, 12] %



54.5 (66)




12.3 (15)



Com A

5.7 (7)



Com P

21.4 (26)



Com A + Com P

1.6 (2)



1.6 (2)


14(Fig. 3a, b)

Com P + A1C

1.6 (2)


Com A + P1

0.8 (1)

A1 precommunicating segment of anterior cerebral artery Com A Anterior communicating artery Com P Posterior communicating artery P1 Posterior precommunicating cerebral artery

Table 2

Comparison between predictive risk of shunt need, blood pressure level, frequency and timing of cerebral ischemia

Number of cerebral ischemia: n patients

Timing of cerebral ischemia after clamping (min)

Blood pressure level (mmHg)a

Predictive risk of cerebral ischemia: n patients



Low: 66 Moderate: 40


0.6 [0. 2–1. 5]


High: 9


15.8 [10–25]


Moderate: 6

a : Difference between blood pressure during carotid clamping and baseline level


In patients with carotid stenosis, pre-operative evaluation of cerebrovascular structure depends more and more on non-invasive vascular imaging modalities such as MR or CT angiography to reduce the morbidity related to conventional angiography, which averages 2% rate [11, 22]. Three dimensional phase contrast MR angiography is known to have a high sensitivity for the flow quantification in intracranial cerebral arteries [15, 17]. Our observations are in line with previous anatomical studies which also confirm the sensitivity for depicting the anatomy of the circle of Willis (Table 2). The present study relates circle of Willis hemodynamic and morphology on pre-operative 3D PC angiograms to immediate transient ischemia during carotid clamping in patients undergoing endarterectomy under locoregional anaesthesia. Statistically significant relation was found between the pre-operative risk and the presence of intraoperative ischemia requiring shunt placement. Concerning the high-risk patients on pre-operative evaluation, our results confirm previous reports in patients with controlateral carotid occlusion [10, 14, 18, 19]. The absence of collateral flow on MRA from the posterior circulation is significantly associated with immediate ischemia.

In the patients who had unilateral stenosis or bilateral stenosis, all high-risk patients (n=5) on pre-operative evaluation (absence of collateral flow in major arterial segment in both anterior and posterior parts of the circle of Willis) had an immediate and severe ischemia independent on blood pressure level. These findings are at the opposite to Lee et al.’s study [10]. They observed ischemia in only one of eight patients with incompleteness in both parts of the circle of Willis. However, the major difference in this study is that the features of anterior and posterior parts of the circle of Willis were categorized according to the vessel size seen on 3D time-of-flight MR angiograms. Contrary to our study in which we evaluated the risk on intravascular flow.

Moreover, the findings observed in the six patients who experienced transient ischemia related to systemic hypotension focuses on the hemodynamic influence during carotid clamping. Consequently, pre-operative evaluation can improve intraoperative management of cerebral perfusion. Significant correlation between intracranial collateral circulation and immediate cerebral ischemia may help prevent brain damage. High-risk patient may benefit from techniques that reduce brain metabolism, such as general anaesthesia with immediate shunt placement to avoid flow-related ischemia [4]. In patients with moderate risk, ischemic vulnerability is dependent on blood pressure stability [13]. We attempt, as recommended by Lawrence et al. to maintain systolic blood pressure at least 20 mmHg above the pre-operative pressure [8].

In conclusion, 3D PC MR angiography is a noninvasive method which provides useful information about blood flow in the circle of Willis. Evaluation of cross-filling from the contralateral carotid artery can significantly predict cerebral ischemia which requires an intraluminal shunt procedure.


  1. 1.
    Alpers BJ, Berry RG, Paddison RM (1959) Anatomical studies of the circle of Willis in normal brains. Arch Neurol Psychiat 81:409–418Google Scholar
  2. 2.
    Doblar DD, Plyushcheva NV, Jordan W, McDowell H (1998) Predicting the effect of carotid occlusion during carotid endarterectomy. Comparing transcranial Doppler measurements and cerebral angiography. Stroke 29:2038–2042PubMedGoogle Scholar
  3. 3.
    Fields NS, Bruetman ME, Weibel J (1965) Collateral circulation of the brain. Williams Wilkins, BaltimoreGoogle Scholar
  4. 4.
    Frawley JE, Hicks RG, Gray LJ, Niesche JW (1996) Carotid endarterectomy without a shunt for symptomatic lesions associated with controlateral severe stenosis or occlusion. J Vasc surg 23:421–427PubMedGoogle Scholar
  5. 5.
    Halsey JH (1992) Risks and benefits of shunting in carotid endarterectomy The International Transcranial Doppler Collaborators. Stroke 23:1583–1587PubMedGoogle Scholar
  6. 6.
    Hoksbergen AWJ, Legemate DA, Ubbink DT, Jacobs MJ (2000) Collateral variations in circle of Willis in atherosclerotic population assessed by means of transcranial color-coded duplex ultrasonography. Stroke 31:1656–1660PubMedGoogle Scholar
  7. 7.
    Kim GE, Cho YP, Lim SM (2002) The anatomy of the circle of Willis as a predictive factor for intra-operative cerebral ischemia (shunt need) during carotid endarterectomy. Neurol Res 24:237–240CrossRefPubMedGoogle Scholar
  8. 8.
    Lawrence PF, Alves JC, Jiche D, Bhirangi K, Dobrin PB (1998) Incidence, timing, and causes of cerebral ischemia during endarterectomy with regional anesthesia. J Vasc Surg 27:329–337PubMedGoogle Scholar
  9. 9.
    Lazorthes G, Gouaze A, Salamon G (1976) Vascularisation et circulation de l’encéphale. Masson, ParisGoogle Scholar
  10. 10.
    Lee JH, Choi CG, Kim DK, Kim GE, Lee HK, Suh DC (2004) Relationship between circle of Willis on 3D Time-of-Flight MR angiograms and transient ischemia during vascular clamping of the internal carotid artery during carotid endarterectomy. Am J Neuroradiol 25:558–564PubMedGoogle Scholar
  11. 11.
    Leffers AM, Wagner A (2000) Neurologic complications of cerebral angiography. A retrospective study of complications rate and patient risk factors. Acta Radiol 41:204–210CrossRefPubMedGoogle Scholar
  12. 12.
    Liebeskind DS (2003) Collateral circulation. Stroke 34:2279CrossRefPubMedGoogle Scholar
  13. 13.
    Mc Carthy RJ, Nasr MK, McAteer P, Horrocks M (2002) Physiological advantages of cerebral blood flow during carotid endarterectomy under local anaesthesia. A randomised clinical trial. Eur J Vasc Endovasc Surg 24:215–221CrossRefPubMedGoogle Scholar
  14. 14.
    Miralles M, Dolz JC, Cottillas J (1995) The role of the circle of Willis in carotid occlusion: assessment with phase contrast MR angiography and transcranial duplex. Eur J Vasc Endovasc Surg. 10:424–430PubMedGoogle Scholar
  15. 15.
    Patrick JT, Fritz JV, Adamo JM, Dandonna P (1996) Phase-contrast magnetic resonance angiography for the determination of cerebrovascular reserve. J Neuroimaging. 6:137–143PubMedGoogle Scholar
  16. 16.
    Pistolese GR, Ippoliti A, Appoloni A, Ronchey S, Faraglia V (1993) Cerebral haemodynamics during carotid cross-clamping. Eur J Vasc Surg 7(supp A):33–38CrossRefPubMedGoogle Scholar
  17. 17.
    Rutgers DR, Blankensteijn JD, van der Grond J (2000) pre-operative MRA flow quantification in CEA patients: flow differences between patients who develop cerebral ischemia and patients who do not develop cerebral ischemia during cross-clamping of the carotid artery. Stroke 31:3021–28PubMedGoogle Scholar
  18. 18.
    Schomer DF, Marks MP, Steinberg GK, Johnstone IM, Boothroyd DB, Ross MR, Pelc NJ, Enzmann DR (1994) The anatomy of the posterior communicating artery as a risk factor for ischemic cerebral infarction. NEJM 330:1565–1570 CrossRefPubMedGoogle Scholar
  19. 19.
    Schwartz RB, Jones KM, Leclercq GT, Ahn SS, Chabot R, Whittemore A, Mannick JA, Donaldson MC, Gugino LD (1992) The value of cerebral angiography in predicting cerebral ischemia during carotid endarterectomy. Am J Roentgenol 159:1057–1061Google Scholar
  20. 20.
    Visser GH, Wieneke GH, van Huffelen AC, Eikelboom BC (2000) The use of pre-operative transcranial Doppler variables to predict which patients do not need a shunt during carotid endarterectomy. Eur J Vasc Endovasc Surg 19:226–232CrossRefPubMedGoogle Scholar
  21. 21.
    Wain RA, Veith FJ, Berkowitz BA, Legatt AD, Schwartz M, Lipitz EC, Haut SR, Bello JA (1999) Angiographic criteria reliably predict when carotid endarterectomy can be safely performed without a shunt. J Am Coll Surg 189:93–100CrossRefPubMedGoogle Scholar
  22. 22.
    Willinsky RA, Taylor SM, TerBrugge K, Farb RJ, Tomlinson G, Montanera W (2003) Neurologic complications of cerebral angiography:prospective analysis of 2899 procedures and review of the litterature. Radiology 227:522–528PubMedGoogle Scholar

Copyright information

© Springer-Verlag 2005

Authors and Affiliations

  • P. Bagan
    • 1
    • 6
  • J. Azorin
    • 1
  • J. Salama
    • 2
    • 4
  • J-L. Dumas
    • 3
    • 4
    • 5
  1. 1.Department of Thoracic and Vascular surgery, Avicenne HospitalAP-HP, University Paris XIIIBobignyFrance
  2. 2.Department of Neurology, Avicenne HospitalAP-HP, University Paris XIIIBobignyFrance
  3. 3.Department of Radiology, Avicenne HospitalAP-HP, University Paris XIIIBobignyFrance
  4. 4.Department of Applied Anatomy of Paris-NordUFR SMBH, University of Paris XIIIBobignyFrance
  5. 5.Department of Applied Anatomy of EAUFR SMBH, University of Paris XIIIBobignyFrance
  6. 6.service de Chirurgie ThoraciqueHôpital Europèen Georges PompidouParisFrance

Personalised recommendations