ORIGINAL ARTICLE

pISSN 2288-6575 • eISSN 2288-6796

https://doi.org/10.4174/astr.2017.92.2.105
Annals of Surgical Treatment and Research

Cerebral monitoring during carotid endarterectomy by

transcranial Doppler ultrasonography
Woo-Sung Yun
Division of Transplantation and Vascular Surgery, Department of Surgery, Yeungnam University Medical Center, Yeungnam
University College of Medicine, Daegu, Korea

Purpose: To evaluate the efficacy and safety of cerebral monitoring by transcranial Doppler ultrasonography (TCD) for the
detection of cerebral ischemia during carotid endarterectomy (CEA).
Methods: From August 2004 to December 2013, 159 CEAs were performed in a tertiary hospital. All procedures were
performed under general anesthesia. Intraoperative TCD was routinely used to detect cerebral ischemia. Of the 159
patients, 102 patients were included in this study, excluding 27 patients who had a poor transtemporal isonation window
and 30 patients who used additional cerebral monitoring systems such as electroencephalography or somatosensory
evoked potentials. When mean flow velocity in the ipsilateral middle cerebral artery decreased by >50% versus baseline
during carotid clamping carotid shunting was selectively performed. The carotid shunt rate and incidence of perioperative
(<30 days) stroke or death were investigated by reviewing medical records.
Results: Carotid shunting was performed in 31 of the 102 patients (30%). Perioperative stroke occurred in 2 patients (2%);
a minor ischemic stroke caused by embolism in one and an intracerebral hemorrhage in the other. Perioperative death
developed in the latter patient.
Conclusion: TCD is a safe cerebral monitoring tool to detect cerebral ischemia during CEA. It can reduce use of carotid
shunt.
[Ann Surg Treat Res 2017;92(2):105-109]

Key Words: Carotid endarterectomy, Transcranial Doppler ultrasonography

INTRODUCTION Selective shunting has theoretical advantages over routine

shunting, because of possible adverse effects, which include
Based on the results of large-scale clinical studies [1-5], carotid intimal damage or dissection, distal embolization, and acute
endarterectomy (CEA) has been recommended for symptomatic occlusion. Furthermore, shunt placement requires more distal
and asymptomatic patients with high-grade carotid artery exposure of the ICA, and this may be challenging in patients
stenosis [6]. Safety has been emphasized during the procedure, with a high lesion. Neurologic monitoring of awake patients
because of the possibilities of disabling or nondisabling stroke. under local or regional anesthesia provides the most reliable
The majority of patients tolerate temporary interruption of means of identifying candidates for selective shunting [8-11],
cerebral blood flow during CEA, but cerebral ischemia may but it cannot be used under general anesthesia. Accordingly,
develop in patients with inadequate collateral channels. Carotid many surgeons that favor general anesthesia adopt routine
shunting guarantees adequate cerebral blood flow during shunting. On the other hand, other surgeons use selective
carotid clamping, but its routine use is controversial [7]. shunting and intraoperative monitoring by transcranial

Received July 8, 2016, Revised September 6, 2016, Copyright ⓒ 2017, the Korean Surgical Society
Accepted September 19, 2016 cc Annals of Surgical Treatment and Research is an Open Access Journal. All
Corresponding Author: Woo-Sung Yun articles are distributed under the terms of the Creative Commons Attribution Non-
Division of Transplantation and Vascular Surgery, Department of Surgery, Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which
Yeungnam University Medical Center, Yeungnam University College of permits unrestricted non-commercial use, distribution, and reproduction in any
Medicine, 170 Hyeonchung-ro, Nam-gu, Daegu 42415, Korea medium, provided the original work is properly cited.
Tel: +82-53-620-3580, Fax: +82-53-624-1213
E-mail: [email protected]

Annals of Surgical Treatment and Research 105

 Annals of Surgical Treatment and Research 2017;92(2):105-109

Doppler ultrasonography (TCD), electroencephalography (EEG), (Fig. 1A). Mean flow velocity (MFV) of the ipsilateral middle
somatosensory evoked potential (SSEP), near infrared spec­ cerebral artery (MCA) was monitored before and after carotid
troscopy, or carotid stump pressure. However, optimal methods clamping (Fig. 1B, C). And the mean blood pressure (MBP)
for detection of cerebral ischemia remain controversial [12]. was also checked at the same time. The criterion for selective
The aim of this study was to evaluate efficacy and safety shunting was a MFV reduction >50% after carotid clamping
of cerebral monitoring by TCD for the detection of cerebral versus baseline.
ischemia during CEA.

METHODS Table 1. Patient demographic data and lesion characteristics

Variable Value
Patients
One hundred fifty-nine patients that underwent CEA Age (yr), mean (range) 66 (43–78)
from August 2004 to December 2013 were retrospectively Male sex 94 (92)
reviewed. All CEA were done under general anesthesia. Intra­ Smoking 50 (49)
operative TCD was routinely used to detect cerebral ischemia. Comorbidities
Hypertension 73 (72)
Of these 159 patients, 27 patients were excluded for a poor
Diabetes mellitus 42 (41)
transtemporal isonation window and 30 because additional Coronary artery disease 11 (11)
cerebral monitoring systems, such as, EEG or SSEP, were used. Hyperlipidemia 34 (33)
Accordingly, 102 patients constituted the study cohort. Patients Indications
with monocular blindness or that had experienced a transient Symptomatic 76 (75)
ischemic attack or ischemic stroke within the previous 6 Asymptomatic 26 (26)
months were regarded as symptomatic. Patient and lesion Lesion site
Right 52 (51)
characteristics are summarized in Table 1. Patient mean age was
Left 50 (49)
66 years (43–78 years), 92% were male, 75% were symptomatic,
Ipsilateral ICA stenosis degree (%)
and 81% had severe (>70%) internal carotid artery (ICA) stenosis. ≥70 83 (81)
Nine patients had contralateral ICA stenosis exceeding 50%. 50–69 19 (19)
Level of stenosis
Carotid endarterectomy C2 6 (6)
All CEAs were performed under general anesthesia with the C2–3 18 (18)
use of nitrous oxide and halothane or isoflurane. Before carotid C3 40 (39)
C3–4 22 (22)
clamping, heparin was administered intravenously; protamine
≤C4 16 (16)
reversal was not used. The eversion technique was used in 56
Ulcerative lesion 26 (26)
patients (55%) and conventional CEA in 46 patients (45%, 36 Contralateral ICA stenosis degree (%)
primary closures and 10 patch closures). ≥70 2 (2)
50–69 7 (7)
TCD monitoring <50 93 (90)
Intraoperative TCD monitoring was performed using a Values are presented as mean (range) or number (%).
TC8080 (Pioneer TC 8080, Nicolet Vascular, Madison, WI, USA) ICA, internal carotid artery.

9:43:46 AM 9:44:44 AM
save 16 (auto): Ch 1A (S1) - 2 MHz PW:MCA save 17 (manual): Ch 1A (S1) - 2 MHz PW:MCA
171 67 171 21
Powet Powet
Sya Sya
50 50
Depth 28 Depth 14

1.1 Dia 0.45 Dia

P1 P1
1 0
35 17
A Mean
B
Embol Mean
C
Embol

Fig. 1. Intraoperative transcranial Doppler (TCD) monitoring. (A) A hands-free standard 2-MHz pulsed wave TCD transducer
was positioned on the ipsilateral temporal bone window with head frame. Mean flow velocity of the ipsilateral middle cerebral
artery (MCA) was measured before (B) and after (C) carotid clamping.

106
 Woo-Sung Yun: Cerebral monitoring during CEA by TCD

Table 2. Change of MFV of ipsilateral MCA after carotid trial failed to identify a definite difference between general
clamping and local anesthesia with respect to the incidences of stroke,
Change Mean preclamp Mean postclamp myocardial infarction, or death [13]. One of the theoretical
No (%)
of MFV MBP (mmHg) MBP (mmHg) benefits of locoregional anesthesia is a lower carotid shunt
Decrease
rate. Reported shunt rates from the awake test range from 4.4%
91%–100% 7 (7) 93 83 to 14% [8-11], and patients with contralateral ICA occlusion
81%–90% 2 (2) 85 86 required carotid shunting more frequently than those with a
71%–80% 4 (4) 78 101 unilateral lesion [9,14]. To reduce shunt rates during general
61%–70% 4 (4) 94 99 anesthesia, several intraoperative monitoring modalities have
51%–60% 14 (14) 79 90 been introduced. TCD is one of these modalities and concerns
41%–50% 4 (4) 95 92
have been expressed regarding its safety and efficacy.
31%–40% 13 (13) 90 88
21%–30% 18 (18) 89 91
TCD cannot determine the presence of cerebral ischemia
11%–20% 12 (12) 86 80 precisely during carotid clamping. Reported values for the
0%–10% 13 (13) 91 97 sensitivity and specificity of TCD for predicting the need
Increase for carotid shunting range from 75% to 83% and from 75% to
1%–10% 6 (6) 88 96 96%, respectively, although these values depend on the cutoff
11%–20% 3 (3) 95 96 used [15,16]. Moritz et al. [17] investigated the accuracies of
21%–30% 1 (1) 79 98
monitoring tools with respect to the detection of cerebral
31%–40% - - -
41%–50% - - - ischemia during CEA. In this previous study, the cutoff value
51%–60% 1 (1) 116 133 for 100% sensitivity for TCD was determined to be a 48%
reduction in ipsilateral MCA MFV versus baseline, the cutoff
MFV, mean flow velocity; MCA, middle cerebral artery; MBP,
mean blood pressure. for 100% specificity was 70%. And the best fit cutoff value was
48%, which provided a sensitivity of 100% and a specificity of
86%, because safety is most important in carotid surgery. In the
Outcomes present study, our shunt criterion was similar to their result.
Shunt rate and the incidences of early postoperative (<30 No patient that did not undergo carotid shunting experienced
days) stroke or death were used to evaluate the efficacy and cerebral ischemia during carotid clamping. Therefore, it would
safety of TCD monitoring. appear that a reduction in ipsilateral MCA MFV of >50% as
determined by TCD is a safe criterion for shunt placement to
RESULTS prevent cerebral ischemia during carotid clamping. However,
the overall shunt rate was 30%, which is greater than previously
Mean preclamp ipsilateral MCA MFV and MBP were 50 cm/ reported shunt rates from the awake test. This elevated rate
sec (21–123 cm/sec) and 88 mmHg (47–130 mmHg), and mean was presumably due to false-positives. If the cutoff value is
postclamp ipsilateral MFV and MBP were 32 cm/sec (0–106 lowered, unnecessary shunting would be avoided. However,
cm/sec) and 91 (60–144 mmHg). Table 2 shows MFV change it will increase a risk of missing cerebral ischemia. Therefore,
after carotid clamping. The mean decrease in ipsilateral MFV to determine an optimal TCD criterion that has lowest shunt
was 33% (-56%–100%). Of the 102 patients, 31 (30%) showed rate and do not make cerebral ischemia during carotid clamp,
an ipsilateral MFV decrease >50% after carotid clamping, additional large-scale clinical studies are necessary.
and carotid shunting was performed in all these patients. Delayed shunt occlusion is a complication of carotid shunting
Accordingly, the shunt rate was 30%. [18-20]. Kink of shunt or adhesion of the shunt stump to the
Regarding early postoperative outcomes, perioperative stroke arterial wall could cause shunt malfunction. Furthermore,
occurred in 2 patients (2%). One minor ischemic stroke caused carotid shunt occlusion would not be recognized in the absence
by embolism occurred in a patient without a carotid shunt. of cerebral ischemia monitoring. If these anomalies are not
The other was due to intracerebral hemorrhage (ICH) caused by corrected immediately, cerebral ischemia may result. On the
hyperperfusion syndrome. One perioperative death developed other hand, continuous TCD monitoring can confirm shunt
in the patient with hyperperfusion syndrome. patency and represents an additional benefit of TCD.
TCD monitoring not only provides information on
DISCUSSION hemodynamic changes but also enables the detection of cerebral
microembolism. Ackerstaff et al. [21] reported the outcomes
Type of anesthesia (general vs . locoregional) remains a of 1,058 patients that underwent CEA with TCD monitoring.
controversial issued for CEA. A large randomized controlled They found 31 patients with ischemic stroke and 8 patients

Annals of Surgical Treatment and Research 107

 Annals of Surgical Treatment and Research 2017;92(2):105-109

with hemorrhagic stroke. TCD-detected microembolism during reduction of ≤50%, carotid clamping could be done without a
dissection and wound closure, ≥90% MCA velocity decrease carotid shunt. Although the technical limitations of TCD were
at carotid clamping and ≥100% increase of pulsatile index (PI) apparent as 10%–15% of patients did not have an adequate
were associated with perioperative stroke or stroke-related temporal window [22], its use during general anesthesia can
death. It was suggested that information on microembolization reduce shunt and shunt-related complication rates.
can caution surgeons regarding they need for more careful
dissection or early carotid clamping and TCD monitoring is a CONFLICTS OF INTEREST
sensitive tool for the prediction of postoperative hyperperfusion
syndrome. In our series, the low sample size prevented our No potential conflict of interest relevant to this article was
analyzing the association between microembolism and stroke. reported.
Regarding hyperperfusion syndrome, the patient that died of
ICH showed an increase in PI of 92% after CEA. ACKNOWLEDGeMENTS
Summarizing, TCD monitoring was found to be a safe
modality for indicating the need for selective shunting during This work was supported by a Yeungnam University Research
CEA. For patients that exhibited am ipsilateral MCA MFV Grant (2015).

REFERENCES

1. North American Symptomatic Carotid JM, Barr JD, Bush RL, et al. 2011 ASA/ u­ation of electroencephalography, carotid
Enda­rterectomy Trial Collaborators. Bene­ ACCF/AHA/AANN/AANS/ACR/ASNR/CNS/ artery stump pressure, and neuro­logic
ficial effect of carotid endarterectomy in SAIP/SCAI/SIR/SNIS/SVM/SVS guideline changes during 314 consecutive carotid
symptomatic patients with high-grade on the management of patients with endarterectomies performed in awake pa­
carotid stenosis. N Engl J Med 1991;325: extra­cranial carotid and vertebral artery tients. J Vasc Surg 2007;45:511-5.
445-53. disease: executive summary. A report 9. Lawrence PF, Alves JC, Jicha D, Bhirangi
2. Endarterectomy for asymptomatic carotid of the American College of Cardiology K, Dobrin PB. Incidence, timing, and
artery stenosis. Executive Committee for Founda­tion/American Heart Association causes of cerebral ischemia during carotid
the Asymptomatic Carotid Atherosclerosis Task Force on Practice Guidelines, and the endarterectomy with regional anesthesia.
Study. JAMA 1995;273:1421-8. American Stroke Association, American J Vasc Surg 1998;27:329-34.
3. Barnett HJ, Taylor DW, Eliasziw M, Fox AJ, Asso­c iation of Neuroscience Nurses, 10. McCarthy RJ, Walker R, McAteer P, Budd
Ferguson GG, Haynes RB, et al. Benefit of Ame­r ican Association of Neurological JS, Horrocks M. Patient and hospital bene­
carotid endarterectomy in patients with Sur­geons, American College of Radiology, fits of local anaesthesia for carotid endar­
symptomatic moderate or severe stenosis. Ame­r ican Society of Neuroradiology, terectomy. Eur J Vasc Endovasc Surg 2001;
North American Symptomatic Carotid Con­g ress of Neurological Surgeons, So­ 22:13-8.
Endarterectomy Trial Collaborators. N ciety of Atherosclerosis Imaging and 11. Stoughton J, Nath RL, Abbott WM.
Engl J Med 1998;339:1415-25. Pre­ven­t ion, Society for Cardiovascular Com­p a­r ison of simultaneous electro­
4. Randomised trial of endarterectomy for Angio­graphy and Interventions, Society ence­­phalographic and mental status
recently symptomatic carotid stenosis: of Interventional Radiology, Society of mo­ni­toring during carotid endarterec­to­
final results of the MRC European Carotid Neuro­Inter­ventional Surgery, Society for my with regional anesthesia. J Vasc Surg
Surgery Trial (ECST). Lancet 1998;351: Vas­cular Medicine, and Society for Vas­ 1998;28:1014-21.
1379-87. cular Surgery. Circulation 2011;124:489- 12. Chongruksut W, Vaniyapong T, Rerkasem
5. Halliday A, Mansfield A, Marro J, Peto 532. K. Routine or selective carotid artery
C, Peto R, Potter J, et al. Prevention of 7. Liapis CD, Bell PR, Mikhailidis D, Sivenius shun­ting for carotid endarterectomy (and
dis­abling and fatal strokes by successful J, Nicolaides A, Fernandes e Fernandes J, dif­ferent methods of monitoring in selec­
caro­tid endarterectomy in patients with­ et al. ESVS guidelines. Invasive treatment tive shunting). Cochrane Database Syst
out recent neurological symptoms: ran­ for carotid stenosis: indications, tech­ni­ Rev 2014;(6):CD000190.
domised controlled trial. Lancet 2004; ques. Eur J Vasc Endovasc Surg 2009;37(4 13. GALA Trial Collaborative Group, Lewis
363:1491-502. Suppl):1-19. SC, Warlow CP, Bodenham AR, Colam B,
6. Brott TG, Halperin JL, Abbara S, Bacharach 8. Hans SS, Jareunpoon O. Prospective eval­ Rothwell PM, et al. General anae­s­the­sia

108
 Woo-Sung Yun: Cerebral monitoring during CEA by TCD

versus local anaesthesia for ca­rotid sur­ Surg 2011;12:454-7. Surg (Torino) 1986;27:146-53.
gery (GALA): a multicentre, ran­domised 17. Moritz S, Kasprzak P, Arlt M, Taeger K, 20. Cho I, Smullens SN, Streletz LJ, Fariello
controlled trial. Lancet 2008;372:2132-42. Metz C. Accuracy of cerebral monitoring RG. The value of intraoperative EEG moni­
14. Hafner CD, Evans WE. Carotid endarter­ in detecting cerebral ischemia during toring during carotid endarterec­tomy.
ectomy with local anesthesia: results and caro­tid endarterectomy: a comparison of Ann Neurol 1986;20:508-12.
advantages. J Vasc Surg 1988;7:232-9. trans­cranial Doppler sonography, near- 21. Ackerst aff RG, Moons KG, van de
15. Cao P, Giordano G, Zannetti S, De Rango P, in­frared spectroscopy, stump pressure, Vlasakker CJ, Moll FL, Vermeulen FE,
Maghini M, Parente B, et al. Transcranial and somatosensory evoked potentials. Algra A, et al. Association of intraoperative
Doppler monitoring during carotid endar­ Anesthesiology 2007;107:563-9. trans­cranial doppler monitoring variables
terectomy: is it appropriate for selecting 18. Artru AA, Strandness DE Jr. Delayed caro­ with stroke from carotid endarterectomy.
patients in need of a shunt? J Vasc Surg tid shunt occlusion detected by electro­en­ Stroke 2000;31:1817-23.
1997;26:973-9. cephalographic monitoring. J Clin Monit 22. Pennekamp CW, Moll FL, de Borst GJ. The
16. Ali AM, Green D, Zayed H, Halawa M, 1989;5:119-22. potential benefits and the role of cerebral
El-Sakka K, Rashid HI. Cerebral moni­ 19. Blackshear WM Jr, Di Carlo V, Seifert KB, monitoring in carotid endarterectomy.
toring in patients undergoing carotid Connar RG. Advantages of continuous Curr Opin Anaesthesiol 2011;24:693-7.
end­arterectomy using a triple assessment elec­tro­encephalographic monitoring du­
tech­n ique. Interact Cardiovasc Thorac ring carotid artery surgery. J Cardiovasc

Annals of Surgical Treatment and Research 109