740

Detection of Paradoxical Cerebral EchoContrast Embolization by

Transcranial Doppler Ultrasound

Steve M. Teague, MD, and Mukesh K. Sharma, MD

Contrast echocardiography has been shown to be a sensitive method for detecting patentforamen ovale in embolic stroke, implying paradoxical embolization. However, not all two-dimensional echocardiographic studies are of diagnostic quality, and direct evidence forparadoxical cerebral embolization remains lacking. We addressed these problems by simulta-neously using transcranial Doppler ultrasound and contrast echocardiography to comparerelative sensitivity and concordance in the detection of right-to-left vascular shunting. Forty-sixpatients with stroke, transient neurologic defect, or question of atrial septal defect underwentstudy at rest and during Valsalva strain. Two-dimensional echocardiography detected shuntingin 26% at rest and 15% during Valsalva strain, whereas transcranial Doppler study returnedrates of 41% and 41%, respectively. Concordance was 82% and 75%, respectively. Discordantstudies almost always had evidence of paradoxical contrast embolization by transcranialDoppler and intermediate findings by two-dimensional echocardiography. Transcranial Dopp-ler is a sensitive, unambiguous technique for the detection of anatomic substrates and targetorgan involvement in patients suspected to have paradoxical cerebral embolization. (Stroke1991^2:740-745)

The etiology of stroke in more than a third ofyoung adult patients may remain uncertaineven after extensive clinical study.1 Paradox-

ical embolization via patent atrial foramen ovale hasbecome suspect as a cause of embolic strokes ortransient ischemic attacks.2"3 Contrast echocardiogra-phy with patients at rest and during provocativemaneuvers has been shown to be a sensitive methodfor noninvasively detecting patent foramen ovale inthese patients.1-3 However, not all two-dimensionalechocardiographic studies are of diagnostic quality.Moreover, the technique only reveals the substratefor paradoxical embolization; the arterial delivery ofemboli originating in the venous system at criticaltarget organs is not demonstrated.

The objective of contrast echocardiography in theevaluation of interatrial shunting is the detection ofmicrocavitation gas bubbles entering left heart cham-bers after intravenous injection. As an alternative toultrasonic imaging, Doppler ultrasound has served asa sensitive detector of blood-borne gas bubbles in

From the University of Oklahoma Health Sciences Center,Cardiology Section, Oklahoma City, Okla.

Presented in part at the 61st Scientific Sessions of the AmericanHeart Association.

Address for correspondence: Steve M. Teague, MD, Metro-Health Medical Center, 3395 Scranton Road, Cleveland, OH44109.

Received August 14, 1990; accepted February 21, 1991.

decompression sickness by demonstrating air bubblesin the pulmonary artery.4-7 Specially designed tran-scranial Doppler devices have been used to detectmicro air emboli in the middle cerebral artery duringsurgery.8-9

The purpose of this study was to use transcranialDoppler ultrasound as a detector of gas bubbles in themiddle cerebral artery simultaneously with two-dimen-sional apical echocardiography as a detector of bubblesin the left atrium and ventricle so as to compare relativesensitivity and demonstrate concordance in the detec-tion of right-to-left vascular shunting.

Subjects and MethodsWe studied 46 consecutive patients referred for

contrast echocardiographic examination in the eval-uation of interatrial shunting. The average age was41 ±7 years, and 21 (46%) were females. Twenty-sixpercent were referred for fixed neurologic deficit,another 26% for transient neurologic deficit, and theremaining 48% were investigated to exclude atrialseptal defect.

We used a 2-MHz pulsed-wave transcranial Dopp-ler device (TC-64e, Carolina Medical Electronics,Inc., King, N.C.) with a spherically focused transducer(sample volume depth, 45-60 mm) to sonicate theright middle cerebral artery (Figure 1). The trans-ducer was placed between the right orbit and external

by guest on August 25, 2016

http://stroke.ahajournals.org/D

ownloaded from

Teague and Sharma Transcranial Doppler and Paradoxical Embolization 741

2DE

FIGURE 1. Schematic representation of simultaneous two-dimensional echocardiographic (2DE) and transcranialDoppler (TCD) studies of cardinal cardiac chambers andmiddle cerebral artery during peripheral venous injection ofecho contrast (EC). The two video images (panels A and B)are superimposed and displayed on a common screen. SV,sample volume.

ear along the zygomatic arch, and its position wasadjusted to find a transtemporal ultrasonic windowallowing the best sonication of the middle cerebralartery along its proximal course.10 A helmet wasfashioned to hold the transducer in position during thestudy. Fast Fourier-analyzed video displays of thereturning Doppler velocity signals were mixed withtwo-dimensional echocardiographic video signals anda voice channel and were continuously displayed andrecorded throughout the testing period.

Microcavitation contrast was generated by agitat-ing a mixture of 0.2 ml of air, 0.5 ml of the patient'sblood, and 5 ml of normal saline between two 10-mlsyringes connected by a tri-port stopcock.1 Bubblesize was measured by immediately filling a microcap-illary tube with the agitated contrast mixture andmeasuring individual bubbles against a calibratedmicroscopic grid. No bubbles smaller than 24 orlarger than 144 fim were observed (Figure 2).

The contrast medium was injected rapidly into aleft antecubital vein during suspended respirationwith or without Valsalva strain. Two-dimensional

200

150

Oo

100

24 48 72 96 120BUBBLE SIZE (microns)

FIGURE 2. Microcavitation size histogram of the saline-blood-air hand-agitated mixture used in this study.

echocardiography was performed with the patient inthe left lateral decubitus position using the apicalfour-chamber view. Simultaneous, continuous two-dimensional echocardiography and transcranialDoppler sonography were performed before, during,and after contrast injection. Appearance times ofcontrast in the left ventricle and middle cerebralartery were recorded.

We considered the two-dimensional echocardio-graphic study positive for interatrial shunting onunequivocal visualization of contrast microcavita-tions in the left atrium or ventricle (Figure 3).Appearance of microbubbles in the cerebral circula-tion was indicated by a "chirp" in the acoustic signalsfrom the middle cerebral artery and a correspondingvideo spike. As other investigators also have ob-served,6-9 these signals and spikes were of muchhigher amplitude than the background Doppler flowsignal and subjectively correlated in intensity with thequantity of bubbles shunting right-to-left on theechocardiogram (Figure 3B).

ResultsWe performed resting studies on all 46 patients, 39

of whom we tested during Valsalva strain as well. Sixof the seven patients not undergoing Valsalva hadunequivocally positive two-dimensional echocardio-grams and transcranial Doppler studies, thus obviat-ing the need for a Valsalva strain, whereas the otherpatient could not perform the maneuver due toneurologic disability.

Two-dimensional echocardiography detected in-teratrial shunting in 12 of the 46 patients (26%) atrest and in six of the 39 patients (15%) duringValsalva strain (Table 1). Transcranial Doppler studyindicated venous-injected contrast in the middlecerebral artery in 19 of the 46 patients (41%) studiedat rest and in 16 of the 39 patients (41%) studiedduring Valsalva strain. Time to left atrial appearancewas 2.6+0.4 seconds at rest and 2.1 ±0.3 seconds withValsalva strain. Time taken for the contrast to appearin the middle cerebral artery after onset of peripheralinjection was 5.9±2.3 seconds at rest and 4.6±2.5seconds during Valsalva strain.

A positive transcranial Doppler study was obtainedin all patients who had evidence of interatrial shunt-ing by two-dimensional echocardiography, with con-cordances of 82% and 75% for resting and Valsalvastudies, respectively. Discordant studies almost al-ways had evidence of paradoxical embolization bytranscranial Doppler and indeterminate findings(Figure 3C) by two-dimensional echocardiography.

DiscussionTranscranial Doppler provided direct evidence of

paradoxical cerebral echo contrast embolization inall patients who had interatrial shunting by two-dimensional echocardiography and resolved thequestion of shunting in those with uninterpretableprecordial studies. The sensitivity of transcranialDoppler was particularly superior during Valsalva

by guest on August 25, 2016

http://stroke.ahajournals.org/D

ownloaded from

742 Stroke Vol 22, No 6 June 1991

by guest on August 25, 2016

http://stroke.ahajournals.org/D

ownloaded from

Teague and Sharma Transcranial Doppler and Paradoxical Embolization 743

FIGURE 3. Representative two-dimensional echocardiograms and transcranial assessments of middle cerebral artery flowvelocity in a patient negative for intracardiac shunting (panel A), positive for intracardiac shunting (panel B), and positive forshunting by transcranial echo but with an uninterpretable two-dimensional echocardiogram (panel C). The appearance ofbubbles in the middle cerebral flow velocity recording is indicated by "B".

straining, during which two-dimensional echocardi-ography usually became uninterpretable, but largequantities of bubbles often appeared in the middlecerebral artery.

Several investigators have studied the usefulness ofcontrast two-dimensional echocardiography for de-tecting patent foramen ovale and interatrial shuntingin young patients presenting with nonhemorrhagiccerebrovascular events. Lechat et al2 studied 60 stroke

TABLE 1. Concordance Between EchocardJographlc andDoppler Studies

At rest (n=46)

2DEchoTCD

During Valsalva strain (n=39)2DEchoTCD

Interatrial

Positive

26% (12)41% (19)

15% (6)41% (16)

shunting

Negative orindeterminate

74% (34)59% (27)

85% (30)59% (23)

Numbers in parentheses indicate total patients in each category.2D Echo, precordial apical four-chamber echocardiography,

TCD, transcranial Doppler ultrasound.

patients less than age 55 and found that 30% hadevidence of interatrial shunting by two-dimensionalechocardiography at rest and an additional 10% dur-ing Valsalva strain. Age-matched control subjects hada 10% incidence of shunting. Similarly, Webster et al3

found that 50% of stroke patients less than 40 years ofage had right-to-left shunts compared to 15% ofage-matched controls. Angiographic correlation ofinteratrial shunting seen by two-dimensional echocar-diography was provided by Harvey et al, who studied11 stroke patients, eight of whom had patent foramenovale or small atrial septal defects confirmed by car-diac catheterization or surgery.

Interatrial shunting is commonly discovered usingcontrast two-dimensional echocardiography in youngpatients who present with nonhemorrhagic cere-brovascular events.1"3 However, contrast echocardi-ography provides evidence only for material passingfrom the venous system to the arterial circulation.Target organ involvement is not demonstrated, andinterpretation of the echocardiographic image is al-ways difficult in patients with poor acoustic charac-teristics. Although this study was not designed toreexplore the prevalence of patent foramen ovale inembolic stroke, it does illustrate the difficulties en-

by guest on August 25, 2016

http://stroke.ahajournals.org/D

ownloaded from

744 Stroke Vol 22, No 6 June 1991

countered in detecting left-to-right contrast shuntingin many echocardiographic studies. The clinical im-plication of this study is that the sensitivity forright-to-left shunt detection can be improved byDoppler monitoring of the cerebral arterial circula-tion after venous contrast delivery.

Passage of the small microbubbles used in thisstudy to the cerebral circulation appeared to be welltolerated. No patient reported symptoms during orafter the contrast study. These observations areconcordant with a survey of contrast echocardiogra-phy safety reported by the American Society ofEchocardiography.11

The air-blood interface is an excellent reflector ofDoppler ultrasound, producing an acoustic ampli-tude many orders of magnitude greater than thereflection from flowing red blood cells. This phenom-enon was first exploited by Gillis and coworkers4 forthe in vivo detection of circulating gas emboli duringdecompression sickness in animals and later to detectair embolism during neurosurgical procedures andcardiopulmonary bypass using the transcranial Dopp-ler Technique.8-9

Transcranial Doppler study may not be possiblewhen the transcranial ultrasonic window or the mid-dle cerebral artery signal cannot be located. How-ever, a difficult window was a problem in only onesubject, whereas six of the two-dimensional echocar-diographic studies were of inadequate diagnosticquality. Even in those with satisfactory two-dimen-sional echocardiographic studies, transcranial Dopp-ler gave less ambiguous evidence for the presence orabsence of shunting.

Transesophageal two-dimensional echocardiogra-phy usually affords excellent visualization of theatrial septum and the foramen ovale.12-14 We haveused this procedure to identify patent foramen ovalein patients during peripheral contrast injection. In-deed, transesophageal contrast echocardiographymay become the reference standard to differentiateshunting across the atrial septum from shuntingacross the pulmonary vascular bed. Transesophagealechocardiography may not be appropriate for manypatients with stroke syndromes, particularly those inwhom Valsalva straining or sedative-induced respira-tory depression might prove deleterious. We reservetransesophageal study to resolve the questions ofpatent foramen ovale or transpulmonic shunting topatients with uninterpretable or impossible precor-dial echocardiographic or transcranial echocardio-graphic examinations.

This study is limited by the absence of normalcontrols and a "reference standard." This study wasdesigned only as a comparison of technique, in whicheach patient served as his or her own control. How-ever, cardiac catheterization appears far less sensi-tive to the presence of patent foramen ovale than thetwo ultrasonic techniques discussed here.115 Onlyone of the major cerebral arteries was sonicated,which may have falsely lowered the sensitivity of thetranscranial Doppler.

Pulmonary shunting is unlikely to account for thepositive studies. The size of microbubbles used in thisstudy was at least four times larger than the red bloodcell, making shunting through pulmonary capillariesvery unlikely. Even if some unusually small mi-crobubbles did pass through intact, the contrasteffect would diminish as transit to the left atrium wasprolonged by traverse of the pulmonary microvascu-lature. Moreover, the time taken for the appearanceof contrast in the left atrium after its injection intothe peripheral vein was 2.1 ±0.4 seconds, whereasappearance after transpulmonic transit appears to begreater than 3 seconds.16 Lastly, echocardiographicshunting of microbubble contrast has been highlycorrelated with the presence of anatomic interatrialdefects, as demonstrated in an earlier study from thislaboratory and by other workers.1-14

We conclude that transcranial Doppler is a sensi-tive, unambiguous technique for the detection ofinteratrial shunting, anatomic substrates, and targetorgan involvement in patients suspected to haveparadoxical embolization.

AcknowledgmentsThe authors wish to thank Paula Logan, RDMS,

and Maureen Tyler for performance of the echocar-diograms, and Barbara Clenney and GlendaHutcheson for manuscript preparation.

References1. Harvey JR, Teague SM, Anderson JL, Voyles WF, Thadani U:

Clinically silent atrial septal defects with evidence for cerebralemboliration. Ann Intern Med 1986;105:695-697

2. Lechat PH, Mas JL, Lascault G, Loron Ph, Theard M,Klimczag M, Drobinski G, Thomas D, Grosgogeat Y: Preva-lence of patient foramen ovale in patients with stroke. N EngiJMed 1988^318:1148-1152

3. Webster MWI, Chancellor AM, Smith HJ, Swift DL, SharpeDN, Bass NM, Glasgow GL: Patent foramen ovale in youngstroke patients. Lancet 1988;2:11-12

4. Gillis MF, Peterson PL, Karagianes MT: In vivo detection ofcirculating gas emboli associated with decompression sicknessusing the Doppler flowmeter. Nature 1968;217:965-966

5. Edmonds-Seal J, Prys-Roberts C, Adams AP: Air embolism: Acomparison of various methods of detection. Anaesthesia 1971;26:202-208

6. Maroon JC, Edmonds-Seal J, Campbell RL: An ultrasonicmethod for detecting air embolism. / Neurosurg 1969;31:196-201

7. Michenfelder JD, Miller RH, Groncrt GA: Evaluation of anultrasonic device (Doppler) for the diagnosis of venous airembolism. Anesthesiology 1972^6:164-167

8. Padayachee TS, Parsons S, Theobold R, Linley J, Gosling RG,Deverall PB: The detection of microemboli in the middlecerebral artery during cardiopulmonary bypass: A transcranialDoppler ultrasound investigation using membrane and bubbleoxygenators. Ann Thome Surg 1987;44:298-302

9. Padayachee TS, Parsons S, Theobold R, Gosling RG, DeverallPB: The effect of arterial filtration on reduction of gaseousmicroemboli in the middle cerebral artery during cardiopul-monary bypass. Ann Thorac Surg 1988;45:647-649

10. Aaslid R (ed): Transcranial Doppler Sonography. New York,Springer-Verlag Inc, 1986, pp 38-51

11. Borruner WJ, Shah P, Allen H, Meltzer R, Kisslo JA: Report ofthe American Society of Echocardiography: Contrast Echocardi-ography. Raleigh, NC, American Society of Echocardiography,1984, pp 1-8

by guest on August 25, 2016

http://stroke.ahajournals.org/D

ownloaded from

Teague and Sharma Transcranial Doppler and Paradoxical Embolization 745

12. Shuaib A: Transesophageal two-dimensional echocardiogra-phy and embolic stroke (letter). Stroke 1988;19:1447

13. Mugge A, Daniel WG, Klopper JW, Lichtlen PR: Visualiza-tion of patent foramen ovale by transesophageal color-codedDoppler echocardiography. Am J Cardiol 1988;62:837-838

14. Zenker G, Erbel R, Kramer G, Mohr-Kahaly S, Drexler M,Harnoncourt K, Meyer J: Transesophageal two-dimensionalechocardiography in young patients with cerebral ischemicevents. Stroke 1988;19:345-348

15. Kronik G, Moesslacher H: Positive contrast echocardiographyin patients with patent foramen ovale and normal right hearthemodynamics. Am J Cardiol 1982;49:1806-1809

16. Berwing K, Schlepper M: Echocardiographic imaging of theleft ventricle by peripheral intravenous injection of echocontrast agent. Am Heart J 1988;115:399-408

KEY WORDS • echocardiography • embolism • ultrasonics

by guest on August 25, 2016

http://stroke.ahajournals.org/D

ownloaded from

S M Teague and M K Sharmaultrasound.

Detection of paradoxical cerebral echo contrast embolization by transcranial Doppler

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 1991 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/01.STR.22.6.740

1991;22:740-745Stroke. 

http://stroke.ahajournals.org/content/22/6/740World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer available in the

Permissions in the middle column of the Web page under Services. Further information about this process isOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Stroke Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on August 25, 2016

http://stroke.ahajournals.org/D

ownloaded from