Eur J Echocardiography 7 Suppl. 2 (2006) S2–S7

Contrast echocardiography: the role ofsulfur hexafluoride in achieving optimal results

Gian Paolo Bezante*, Nadia Girardi, Sergio Agosti, Antonio Barsotti

Cardiovascular Unit, Department of Internal Medicine, University of Genova, Genova, Italy

KEYWORDSReal-time contrast

echocardiography;Cardiac function;Myocardial perfusion.

Abstract Recent advances in contrast echocardiography have improved the non-invasive assessment of cardiac structure and function. The use of harmonic imagingin combination with ultrasonic contrast agents such as SonoVue® provides diagnosticinformation that was previously difficult or impossible to obtain by standard two-dimensional echocardiography. A large number of studies have shown the advantagesof using an ultrasonic contrast agent during stress echocardiography and formyocardial perfusion imaging. We will review the main advantages of using ultrasoniccontrast agents during echocardiographic examination.© 2006 The European Society of Cardiology. Published by Elsevier Ltd. All rightsreserved.

Background

Good definition of the endocardial border (EB) iscrucial for accurate assessment of left ventricularfunction. However, when using standard echotechniques limitations are encountered in 10–20%of patients 1,2. The major limitation is the poorangle of incidence between ultrasound beam andthe endocardium in some areas. Other limitationsare due to habitus and/or anatomy of the patient:(1) obese patients with a poor acoustic window;(2) patients with cardiac displacement; (3) patientswith narrow intercostal spaces; (4) patients withpulmonary disease where intervening lung tissueleads to attenuation of the signal.

The introduction of second harmonic imaging andmicrobubble based ultrasonic contrast agents hasresulted in a better delineation of the endocardialborder. After venous injection, microbubblesremain in the intravascular space, behave as redblood cells and do not impair blood flow. When

* Corresponding author. Gian Paolo Bezante. Cardiovas-cular Unit, Department of Internal Medicine, University ofGenova, Viale Benedetto XV/6, Genova 16132, Italy. E-mailaddress: gpb001@cardio.dimi.unige.it (G.P. Bezante).

microbubbles reach the LV cavity they producea signal increase of 20–40 dB of the blood poolindependently of the ultrasound beam interroga-tion angle (they are scatterers) and, what is mostpeculiar, they sing with a harmonic tune. Indeed,when insonated at a certain acoustic power, anysingle bubble acts as a harmonic oscillator andexhibits nonlinear vibrations. The nonlinearity ofthe contrast produces a “signature” that canbe separated from echoes coming from tissuesand large vessel blood flow, allowing improvedendocardial border delineation and visualisationof capillary blood flow (i.e., perfusion). SonoVue®

(Bracco) is one of the new ultrasound contrastagents, made of stabilized microbubbles containingsulfur hexafluoride (SF6), a highly echogenic, poorlysoluble and totally innocuous gas, with outstandingstability and resistance to pressure 3.

When using microbubbles, a contrast-specificimaging technique is required which is basedon the nonlinear properties of these contrastagents. This technique allows selection of theacoustic power (low or high), transmission of theultrasound (US) signal at fundamental frequencyand its reception at the second harmonic, showing

1525-2167/$30 © 2006 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.

Contrast echocardiography: the role of sulfur hexafluoride in achieving optimal results S3

Fig. 1. Endocardial border delineation pre- and post-US contrast injection. Left: pre-contrast 2-chamber (top) and 4-chamber(bottom) views. Right: post-contrast 2-chamber (top) and 4-chamber (bottom) views.

not only a very high sensitivity to the contrastagent but also high spatial resolution similar tothat of conventional B mode when used in the sametransmission frequency band. Intermittent imagingwith high acoustic output utilizes the uniqueproperty of contrast microbubbles to improveblood-to-tissue signal, producing images whosecontrast emphasizes regions with rapid blood flowor regions with high or low blood volume.

Therefore, the use of an ultrasound contrastagent together with a specific imaging techniqueallows enhancement of the LV blood pool cavityand hence better identification of the endocardialborder 4 (Fig. 1) by separation of the wall signal

Fig. 2. Contrast-specific imaging techniques to separate the wallsignal from the LV cavity flow (adapted from JL Zamorano andMA Garcia Fernandez (editors), Contrast Echocardiography inClinical Practice. 2004; Springer, Milan).

from the LV cavity flow 5,6 (Fig. 2). The contrast-enhanced harmonic imaging produces images withvivid LV cavity opacification without significantartifacts 7–10, providing reasonably accurate mea-surements of the LVEF and LV volumes in patientswith a wide range of image quality and cardiacabnormalities 8,11.

Imaging modalities

In the last few years pulse inversion imaging, powermodulation, or a combination of both (CPS) havebecome available and improve the performanceof echocardiography 12. These techniques alsoimprove the differentiation of contrast from tissue.By modifying the phase (pulse inversion) oramplitude (power modulation) of sequential pulsesof ultrasound sent along the same scan line andsummation of the received signal, the ultrasoundpulses returning from the tissue cancel eachother, while those from the contrast whichproduce higher harmonics, are enhanced. Thesemethods are so sensitive that weaker echoes frommicrobubbles insonated at very low intensities canbe readily imaged, which results in oscillationwithout bubble destruction and hence prolongationof the contrast effect. This principle is the basis ofreal time myocardial perfusion imaging. Standardclinical echocardiography utilises a mechanicalindex (MI) of around 1.0, but a lower setting,around 0.5, is usually optimal for left ventricular

S4 G.P. Bezante et al.

Table 1Clinical indications for right heart opacification by US contrast agents (hand-agitated solutions)

Indication Evaluation

Intracardiac shunt (atrial septal defect, ventricularseptal defect, etc.)

Negative contrast effect: void in opacified RA or RVPassage of contrast from Right to Left Heart

Extracardiac shunt:

patent ductus arteriosus Visualisation of contrast effect in aorta and peripheral circulation withoutleft heart effect

pulmonary arteriovenous fistula Delayed right-to-left shunt with contrast appearance in LA via pulmonaryveins after a delay of 5–15 cardiac cycles

Patent foramen ovale Contrast in LA immediately after visualisation in RA at rest condition,during cough and after Valsalva manoeuvre

Structure identification (persistent left superiorvena cava)

Visualisation of contrast in enlarged coronary sinus and RA after injectioninto left antecubital vein

Evaluation of tricuspid regurgitation Enhanced colour and spectral Doppler

Evaluation of complex congenital heart disease Combination of the above

opacification. In myocardial perfusion imaging, theextremes of power output or MI are utilised:high MI (>1.2) is used to achieve bubble destructionin power Doppler imaging, ultra-low MI (<0.1)induces linear oscillation of microbubbles andis required for real-time myocardial perfusionimaging (Table 1).

The most important clinical applications of UScontrast agents include left ventricular function as-sessment, stress echocardiography and myocardialperfusion imaging (Table 2).

Table 2Clinical indications for left heart opacification by contrastagents

Use Indication

Accurate assessment ofsystolic LV function

Suboptimal imaging, i.e. ventilatedpatientAutomated edge detection

Recognition of regionalwall motion abnormalities

Stress echocardiography

Confirm/exclude LV mass Suspected apical filling defect

Delineate LV structure PseudoaneurysmApical hypertrophic cardiomyopathyNon-compaction of left ventricleIntracardiac mass (vascularity)

Endocardial border delineation/Leftventricular function assessment at restand during stress tests

Several studies have demonstrated the utility ofUS contrast agents for endocardial border delin-eation 13,14. In a multicenter study Nanda et al. 15

showed that SonoVue improves visualization of theendocardial borders in patients with suboptimalnon-contrast echocardiograms and duration ofuseful contrast effect.

In a more recent paper 16 the accuracy andreproducibility of contrast echocardiography versustissue harmonic imaging for measurements of leftventricular (LV) volumes and ejection fraction (EF)were evaluated and magnetic resonance imag-ing (MRI) was used as a reference. Two-dimensionalechocardiographic evaluation of LV volumes andEF in non-selected cardiac patients was foundto be more accurate and reproducible when anintravenous contrast agent was used (Table 3).

At intra-observer analysis of the variability ofEF determination, the limits of agreement were–11.1% to 7.8% pre-contrast, improving to –2.8% to2.4% with contrast. Mean inter-observer variabilitywent from 13.9% to 9.6%. Mean intra-observervariability improved from 5.4% to 2.5%. However

Table 3Volume and ejection fraction pre and post-US contrast. Comparison with magnetic resonancea

N = 100 EDV (ml) ESV (ml) EF (%)

Reference (MRI) 177.0±60.5 (90.3–395.0) 78.7±56.4 (22.9–298.1) 59±14.6 (21–78)Baseline US 126.1±52.2 (48.7–324) 63.0±43.8 (9.5–227.3) 54±12 (18–80)Contrast US 152.2±55.1 (80.8–360.7) 71.1±48.7 (18.1–252.5) 57±13.3 (22–79)

a After Malm et al. (2004) 16.

Contrast echocardiography: the role of sulfur hexafluoride in achieving optimal results S5

Table 4Differences in volume and ejection fraction pre- and post-US contrast between MRI and echocardiography, in relation tobaseline image quality a

N = 87 EDV (ml) ESV (ml) EF (%)

Patients with poor baseline image quality

Without contrast –56.4±47.8 –16.2±31.8 –6.0±13.9

With contrast –24.4±23.8 –6.5±15.6 –1.9±6.5

Patients with good baseline image quality

Without contrast –56.3±41.8 –21.8±47.2 –3.8±12.9

With contrast –25.7±24.4 –9.0±20.2 –1.4±5.9

a After Malm et al. (2004) 16.

what is relevant in this study is that the same istrue also in patients with good images (Table 4).

In another recent multicenter study for theassessment of systolic LV function 17, unenhancedand contrast-enhanced echocardiography was com-pared to cineventriculography and cardiac MRI in120 patients. Contrast echocardiography signifi-cantly improves inter-observer agreement of EF ascompared to conventional imaging and reaches alevel that is comparable to MRI and higher thancineventriculography.

Analysis of regional wall motion abnormalities isalso characterized by considerable inter-observervariability, even when using high-quality imag-ing modalities. (i.e. cineventriculography, MRI).Contrast-enhanced echocardiography as comparedto conventional echocardiography significantly im-proves inter-observer agreement and inter-methodagreement for detecting regional wall motionabnormalities 18.

The use of US contrast is also helpful indemonstrating and delineating LV masses, as shownby our group where excellent agreement with theMRI technique was obtained 19.

In summary, the use of US contrast reducesunderestimation of cardiac volumes and yieldsa better reproducibility of EF measurement,increases concordance between different tech-niques, and reduces inter- and intra-observervariability. However, some limitations of theultrasound technique, such as wrong alignment ofimaging planes, long axis dislocation, geometricalassumptions and cardiac translation, are notovercome by contrast echo.

Good endocardial definition is particularly cru-cial when echocardiography is performed underphysiological or pharmacological stress: the use ofcontrast agents in this situation has been foundto be particularly valuable. In about 35% of stressechocardiographic studies, some segments of the

Table 5Situations in which the use of contrast agents in routine stress-echocardiography can reduce the incidence of false negative andpositive results

False negative False positive

Sub-maximal stress Overestimation, reading error

Medical therapy Postero-inferior wall motion

Non-critical coronary stenosis Left bundle branch block (LBBB)

Circumflex diseases Myocardiopathies

Poor imaging window Hypertensive response to stress

endocardium are poorly seen, and the use ofcontrast agents can render most of these segmentsinterpretable 20. Overall, such agents improve thediagnostic reliability and accuracy and reduceinter-observer variability.

There are disadvantages in the stress-echoprocedure (Table 5), related to poor image qualityand both false positive and false negative resultsmay occur. CE can help avoid these errors. Intheir study on the assessment of myocardialviability using different imaging techniques, Baxet al. 21 showed that low-dose dobutamine stressechocardiography has a sensitivity and specificitysimilar to or better than other imaging techniques.The use of US contrast during stress echofurther increases the sensitivity and specificity byimproving image quality. In another study, by Brownet al. 22, US contrast (SonoVue) administrationwas shown to significantly reduce intra-observerinterpretation variability of end-diastolic, end-systolic volumes and ejection fraction at rest andduring stress by non skilled operators.

Myocardial perfusion

In the last decade, intravenous myocardial contrastechocardiography (MCE) for the assessment ofmyocardial perfusion has become a reality. Thisis because of development of novel approachesto imaging microbubbles. These include high-power modalities with superior tissue noisesuppression that allow on-line assessment ofmyocardial perfusion, and low-power modalitiesthat permit imaging with high frame rates 23–25.These modalities specifically take advantage ofthe non-linear behaviour of microbubbles in anacoustic field compared to myocardial tissue whichmostly has linear behaviour.

One of the first studies assessing MCE with high-power triggered imaging in stable CAD patientsshowed that MCE can define the presence of ab-normal perfusion at rest and during pharmacologicstress, with a high concordance between MCE and99mTc-sestamibi-SPECT of approximately 90% 26.

S6 G.P. Bezante et al.

Several other studies assessed the accuracy of MCEand SPECT for detection of stable CAD. Most reporta sensitivity of MCE for CAD detection that is similarto or higher than SPECT, ranging from 75% to 97%,compared to 43–100% for the latter 27–31. Allthese studies used vasodilator stress. However,MCE has also been shown to improve sensitivityover all motion analysis during dobutamine andexercise stress echocardiography 32,33. Thus, MCE-determined myocardial perfusion accurately de-tects flow-limiting CAD with both vasodilator andinotropic stress.

Application of this technique extends beyondthe detection of CAD. MCE is capable of assessingmyocardial viability, determine infarction size andevaluate reflow and no- or low-reflow after acutemyocardial infarction 34–36.

Conclusion

Contrast echocardiography has evolved rapidly inthe last decade, with major developments in bothcontrast media and ultrasound equipment. At thesame time, understanding the physical principlesunderlying the interaction of ultrasound withmicrobubbles has improved our ability to optimisethe technique, making it more accurate andreproducible. Contrast echocardiography whichwas used predominantly for the detection ofintracardiac shunts is now routinely used forleft heart opacification to optimise assessmentof left ventricular function during stress echocar-diography when imaging is suboptimal. Contrastenhancement should be considered not only invery difficult-to-image patients, but anytime it isdeemed important to have precise and repeatablemeasurements of LV size and global systolicperformance. Further refinement in technology hasbrought us to the threshold of myocardial contrastechocardiography as a tool for the routine clinicalassessment of myocardial perfusion.

References

1. Senior R. Left ventricular contrast echocardiography: rolefor evaluation of function and structure. Echocardiography2002;19:615–20.

2. Lang RM, Mor-Avi V, Zoghbi WA, Senior R, Klein AL,Pearlman AS. The role of contrast enhancement inechocardiographic assessment of left ventricular functionAm J Cardiol 2002;90(Suppl 1):28–34.

3. Schneider M, Arditi M, Barrau MB, Brochot J, Broillet A,Ventrone R, et al. BR1: a new ultrasonographic contrastagent based on sulphur hexafluoride-filled microbubbles.Invest Radiol 1995;30:451–7.

4. Bezante GP, Rosa GM, Bruni R, Chen X, Villa G, Scopinaro A,et al. Improved assessment of left ventricular volumesand ejection fraction by contrast enhanced harmonic

color Doppler echocardiography. Int J Cardiovasc Imag2005;21:609–16.

5. Nanda NC. Clinical utility of the transpulmonary echoenhancing agent Levovist to improve diagnostic confidencein echocardiography. J Clin Ultrasound 1997;25:411–9.

6. Senior R, Andersson O, Caidahl K, Carlens P, Herregods MC,Jenni R, et al. Enhanced left ventricular endocardial borderdelineation with an intravenous injection of Sonovue, a newechocardiographic contrast agent: a European multicentrestudy. Echocardiography 2000;17(8):705–11.

7. Zotz RJ, Genth S, Waaler A, Erbel R, Meyer J. Leftventricular volume determination using Albunex. J Am SocEchocardiogr 1996;9:1–8.

8. Hundley WG, Kizilbash AM, Afridi I, Franco F, Peshock RM,Grayburn PA. Administration of an intravenous per-fluorocarbon contrast agent improves echocardiographicdetermination of left ventricular volumes and ejectionfraction: comparison with cine magnetic resonance imaging.J Am Coll Cardiol 1998;32:1426–32.

9. Colon Jr O, Richards DR, Moreno CA, Murgo JP, Cheirif J.Benefits of reducing the cardiac cycle-triggering frequencyof ultrasound imaging to increase myocardial opacificationwith FS069 during fundamental and second harmonicimaging. J Am Soc Echocardogr 1997;10:602–7.

10. Nahar T, Croft L, Shapiro R, Fruchtman S, Diamond J,Henzlova M, et al. Comparison of four echocardiographictechniques for measuring left ventricular ejection fraction.Am J Cardiol 2000;86:1358–62.

11. Yu EH, Sloggett CE, Iwanochko RM, Rakowski H, Siu S.Feasibility and accuracy of left ventricular volumesand ejection fraction determined by fundamental tissueharmonic and intravenous contrast imaging in difficult-to-image patients. J Am Soc Echocardiogr 2000;13:216–24.

12. Serra V, Fernandez MA; Zamorano JL. Microbubbles: basicprinciples. In: JL Zamorano, MA Fernandez (editors),Contrast Echocardiography in Clinical Practice. 2004;Springer, ch. 2, pp. 19–43.

13. Becher H. Contrast echocardiography: clinical applicationsand future prospects Herz 2002;27(3):201–16.

14. Spencer KT, Bednarz J, Mor-Avi V, Weinert L, Tan J, Godoy I,et al. The role of echocardiographic harmonic imagingand contrast enhancement for improvement of endocardialborder delineation. J Am Soc Echocardiogr 2000;13:131–8.

15. Nanda NC, Wistran DC, Karlsberg RP, Hack TC, Smith WB,Foley DA, et al. Multicenter evaluation of SonoVue forimproved endocardial border delineation. Echocardiography2002;19(1):27–36.

16. Malm S, Frigstad S, Sagberg E, Larsson H, Skjaerpe T.Accurate and reproducible measurement of left ventricularvolume and ejection fraction by contrast echocardiography:a comparison with magnetic resonance imaging. J Am CollCardiol 2004;44(5):1030–5.

17. Hoffmann R, von Bardeleben S, Ten Cate F, Borges AC,Kasprzak J, Firschke C, et al. The use of contrastechocardiography: a matter of clinical judgement. EurHeart J 2005;26(15):1565–6.

18. Hoffmann R, von Bardeleben S, Kasprzak JD, Borges AC,ten Cate F, Firschke C, et al. Analysis of regionalleft ventricular function by cineventriculography, cardiacmagnetic resonance imaging, and unenhanced and contrast-enhanced echocardiography: a multicenter comparison ofmethods. J Am Coll Cardiol 2006;47(1):121–8.

19. Bezante GP, Chen X, Molinari G, Valbusa A, Deferrari L,Sebastiani V, et al. Left ventricular myocardial massdetermination by contrast enhanced colour Dopplercompared with magnetic resonance imaging. Heart2005;91(1):38–43.

Contrast echocardiography: the role of sulfur hexafluoride in achieving optimal results S7

20. Cheng S-C, Dy TC, Feinstein SB. Contrast echocardiography:review and future directions. Am J Cardiol 1998;81(12A):41G–48G.

21. Bax JJ, Wijns W, Cornel JH, Visser FC, Boersma E,Fioretti PM. Accuracy of currently available techniques forprediction of functional recovery after revascularization inpatients with left ventricular dysfunction due to chroniccoronary artery disease: comparison of pooled data. J AmColl Cardiol 1997;30(6):1451–60.

22. Brown AS, Calachanis M, Evdoridis C, Hancock J, Wild S,Prasan A, et al. Sonovue improves endocardial borderdetection and variability in assessing wall motion score andejection fraction during stress echocardiography Ir J MedSci 2004;173(1):13–7.

23. Porter TR, Xie F. Transient myocardial contrast afterinitial exposure to diagnostic ultrasound pressures withminute doses of intravenously injected microbubbles:demonstration and potential mechanisms. Circulation1995;92:2391–5.

24. Senior R, Kaul S, Soman P, Lahiri A. Power doppler harmonicimaging: A feasibility study of a new technique for theassessment of myocardial perfusion Am Heart J 2000 Feb;139(2 pt 1):245–51.

25. Tiemann K, Lohmeier S, Kuntz S, et al. Real-time contrastecho assessment of myocardial perfusion at low emissionpower: first experimental and clinical results using powerpulse inversion imaging. Echocardiography 1999;16:799–809.

26. Kaul S, Senior R, Dittrich H, et al. Detection of coronaryartery disease with myocardial contrast echocardiography:comparison with 99mTc-sestamibi single-photon emissioncomputed tomography. Circulation 1997;96:785–92.

27. Senior R, Lepper W, Pasquet A, et al. Myocardialperfusion assessment in patients with medium probability ofcoronary artery disease and no prior myocardial infarction:comparison of myocardial contrast echocardiography with99mTc single-photon emission computed tomography. AmHeart J 2004;147:1100–5.

28. Wei K, Crouse L, Weiss J, et al. Comparison ofusefulness of dipyridamole stress myocardial contrast

echocardiography to technetium-99m sestamibi single-photon emission computed tomography for detection ofcoronary artery disease (PB127 Multicenter Phase 2 Trialresults). Am J Cardiol 2003;91:1293–8.

29. Peltier M, Vancraeynest D, Pasquet A, et al. Assessment ofthe physiologic significance of coronary disease with dipyri-damole real-time myocardial contrast echocardiography.Comparison with technetium-99m sestamibi single-protonemission computed tomography and quantitative coronaryangiography. J Am Coll Cardiol 2004;43:257–84.

30. Senior R, Janardhanan R, Jeetley P, et al. Myocardialcontrast echocardiography for distinguishing ischemicfrom nonischemic first-onset acute heart failure: insightsinto the mechanism of acute heart failure. Circulation2005;112:1587–93.

31. Jeetley P, Hickman M, Kamp O, et al. Myocardial contrastechocardiography for the detection of coronary arterystenosis: a prospective multicenter study in comparisonwith single-photon emission computed tomography. J AmColl Cardiol. 2006;47:141–5.

32. Xie F, Tsytsyu JM, McGrain AC, et al. Comparison ofdobutamine stress echocardiography with and without real-time perfusion imaging for detection of coronary arterydisease. Am J Cardiol. 2005;96:506–11.

33. Moir S, Halsuka Baa, Jenkins C, et al. Incremental benefitof myocardial contrast to combined dipyridamole–exercisestress echocardiography for the assessment of coronaryartery disease. Circulation 2004;110:1108–13.

34. Hayat SA, Senior R. Contrast echocardiography for theassessment of myocardial viability. Curr Opin Cardiol.2006;21:473–8.

35. Janardhanan R, Moon JCC, Pennell DJ, Senior R. Myocardialcontrast echocardiography accurately reflects transmuralityof myocardial necrosis and predicts contractile reserve afteracute myocardial infarction. Am Heart J 2005;149:355–62.

36. Ito H, Tomooka T, Sakai N, et al. Lack of myocardial perfusionimmediately after successful thrombolysis. A predictorof poor recovery of left ventricular function in anteriormyocardial infarction. Circulation 1992;85:1699–705.