Left ventricular volume determination using Albunex

ORIGINAL ARTICLES

Left Ventricular Volume Determination Using Albunex

Rainer I. Zotz, MD, Sabine Genth, MD, Anne Waaler, DPharm, Raimund Erbel, MD, and Jtirgen Meyer, MD, Mainz, Germany

The aim of this study was to assess the ability of intrave- nous injection o f Albunex to improve left ventricular volume determination in patients with various cardiac diseases. I t is conceivable that the intravenous injection of microbubbles could improve echocardiographic left ventricular border delineation leading to improved in- terobserver variability. Echocardiograms were obtained during simultaneous intravenous injection of 0.08 and 0.12 m l / k g Albunex (four-chamber view). Within 6 hours after the recording of the contrast echocardio- grams routine left heart angiography was performed. Volumes were measured using the slice method in native and contrast echocardiograms as well as in the angio- grams by two independent investigators. The mean dif-

ferences o f angiographic-native echocardiographic and angiographic-contrast echocardiographie volumes (ml) as well as the calculated ejection fraction and their con- fidence intervals were tabulated and significance was an ticipated if the confidence interval did not include zero. Significant changes to angiographic values could be ob- served concerning end-diastolic, echocardiographic, and contrast echocardiographic volumes at end diastole and end systole, while ejection fractions were similar. Differences between observers were significantly smaller in the contrast echocardiographic images than in the native echocardiographic images, k]bunex led to a significantly decreased interobserver variability. (J AM SOC ECHOCARDIOGR 1996;9:1-8.)

In conventional echocardiograms there are often so-called " d r o p - o u t s " (i.c., regions o f left ventricular myocard ium that do no t reflect or scatter ul trasound beams back to the receiver). Consequent ly there is large interobserver variation in left ventricular vol- ume mcasurements ) "2 To delineate the endocar- dium, especially in thcse regions o f drop-outs , intra- venous injection o f contrast material has been pro- posed. Thc aim o f this study was thus to evaluate whether this me thod leads to a morc accurate vol- ume determinat ion and decreased interobserver vari- ability.

METHODS

Patients

Experiments were conducted in accordance with the ethi- cal standards of the Regional Committee on Human Ex- perimentation and with the Declaration of Helsinld and were accepted by the ethics committee of Freiburg, Ger- many. Patients were informed about the study and its risks and if they agreed were consecutively enrolled into the study. The study was composed of 50 consecutive patients, 45 men and 5 women, who were undergoing catheteriza- tion for the evaluation of clinically suspected coronary

From the II Medical Clinic, Johannes Gutenberg University, Mainz, Germany, and Nycomed AS, Oslo, Norway. Reprint requests: Rainer J. Zotz, MD, Heart Center Leipzig University, Russenstr. 19, 04289 Leipzig, Germany. Copyright © 1996 by the American Society of Echocardiography. 0894-7317/96 $5.00 + 0 27/1/684=82

artery disease. The mean age (+SD) was 55 _+ 10 (range 35 to 75) years. Of these patients, 29 had coronary artery disease (1 with severe aortic insufficiency, 1 with severe mitral stenosis and insufficiency, and one with significant coronary narrowing over a length of 3 cm due to myocar- dial bridging).

Subsequently, 18 patients were cathecerized 1 to 6 months after successful percutaneous transluminal coro- nary angioplasty as part of a routine follow-up. Seventeen patients had asynergy of the anterior myocardial wall, five had asynergy of the inferior wall, and one had asynergy of both the anterior and inferior walls.

Exclusion criteria for this study were as follows: age less than 18 or more than 70 years, pregnancy or inadequate contraception or breast feeding women, blood product allergy, protein allergy, HIV or hepatitis infection, sepsis, adult respiratory distress syndrome, medication with non- steroidal antlinflammatory drugs, progressive organ dys- function, mental incompetence, heart transplantation planned or performed, end-stage heart failure, unstable circulatory function, and concomitant investigational drugs within the last 60 days.

Clinical Moni tor ing

Before, immediately after, and 24 hours after injection, a thorough clinical examination was performed. This in- cluded the measurement of systolic and diastolic blood pressure, pulse rate, respiratory rate, oxygen saturation, and temperature.

Two-dimensional Echocardiography

The cross-sectional echocardiographic studies were per- formed with a Toshiba SSH 160A real-time phased array

Journal of the American Society of Echocardiography 2 Z o t z et al. January-February 1996

sector scanner (Toshiba, Tokyo, Japan). The radius of the 2.25 MHz transducer was 0.6 cm at the site ofsldn con- tact. An 84 ° sector could be visualized and we selected a depth of 15 cm. Thirty flames/see were recorded on a video tape recorder (JVC BR-6400U). Real-time and slow motion replays were qualitatively evaluated on a standard television monitor. Echocardiographic images made from an apical transducer position were analyzed. From the region of the apex of the heart, the long axis was trans- sected, providing images of the left ventricle similar to those from the cineventriculographic right anterior ob- lique projection. 1'2 Echoventriculography was performed before, during, and after the intravenous injection of the echo contrast agent. Sufficient echocardiograms were ob- tained in 50 patients in the supine position.

Contrast Agent

Albunex (Molecular Biosystems, San Diego, Calif.) injec- tion consists of air-filled microsphcres of heat aggregated human albumin suspended in Human Albumin Solution 5% (Ph. Eur.). The concentration was 200 million air-filled microspheres/ml with diameters of 4 to 10 btm. The prod- uct is isotonic and has a pH of 6.7 to 7.3. The human albumin is prepared from plasma collected and healthy donors who have been tested and found negative for hepa- titis B surface antigen and antibodies to hepatitis C and HIV. In addition, the albumin solution used for produc- tion of Mbunex injection has undergone virus inactivation by heat treatment at 60 ° C for 10 hours. Albunex injection is a sterile, pyrogen-free suspension. The product in the nonresuspended form is a clear, light yellow to amber solution with a white layer on top. After resuspension an opaque, white to pale yellow, homogenous suspension is obtained. Preclinical animal safety and toxicity studies were conducted to predict the potential safety and efficacy of Albunex in human beings.

Catheterization

Within 6 hours after the recording of the contrast echocar- diograms, the left ventricle was catheterized retrogradely through a percutaneous puncture of the right femoral artery. The patients were fasting and not premedicated. In the 30 ° right anterior oblique projection, 35-mm cineven- triculograms of the left ventricle were exposed at 50 ftames/sec with a contrast medium injection rate of 12 ml/sec using a 0.5 ml/kg dose of meglumine sodium ioxaleglat (BYK-Gulden, Konstanz, Germany). During cineventriculography, the electrocardiogram and frame sig- nals were recorded at a paper speed of 50 mm/sec. No patient was in atrial fibrillation.

End diastole was identified from the Q wave of the electrocardiogram and end systole from the frame with the smallest ventricular silhouette. The magnification factor was assessed by comparison with a metal ball filmed after catheterization in the position of the heart. Left ventricular pressure was recorded through a fluid-filled catheter using a Statham P23 ID transducer (Siemens AG, Erlangen, Germany). Selective coronary cineangiography was per- formed by the Judldns technique. Ventriculograms were reviewed by one experienced angiographer.

Volume Calculations

End-diastolic and end-systolic volumes were analyzed in the native and in the contrast echocardiographic images as well as in cineventriculograms by two independent inves- tigators using the same semiautomatic computer system by means of the slice method. 1'2 The validity of the slice method has been confirmed in asymmetric model hearts for two-dimensional echocardiography and radiography. 2 In all the patients the first three beats after the injection of contrast medium could be analyzed with both methods without extra systoles interrupting the sequence.

Endocardial Contour Finding

The left ventricle was divided into five segments, the first being the lower septal part, the second being the upper septum, the third representing the apex, the fourth the upper posterior wall, and the fifth the lower posterior wall. The number of segments not visualized before injection was compared with the number of segments not visualized after the second injection of Mbunex.

Laboratory Parameters

Values for hemoglobin, hematocrit, numbers of erythro- cytes, leukocytes, neutrophils, bands, segments, basophils, lymphocytes, monocytes, platelets and levels of serum cal- cium, sodium, potassium, total protein, CK, uric acid, urea, total cholesterol, glucose, SGOT, SGPT, GGT and alkaline phosphatase were determined before, 10 minutes after, and 24 hours after the study.

Electrocardiography

Heart rate, P interval, PQ interval, QRS interval, and QT interval were measured in the electrocardiograms taken before, 10 minutes after, and 24 hours after the study.

Statistics

Values before injection of the agents (prevalues) and after injection (postvalues) were calculated and tabulated for each variable. This was done for each agent and each variable early and late after injection. The mean difference of measurements was calculated for each variable as

Echo (N or C) 1/ 100% / o

Anglo

where N = native and C = contrast echocardiographic (Echo) and Anglo = angiographic values. The 95% confi- dence intervals of the mean differences for each of the variables were calculated. If the 95% confidence intervals did not include zero, then the mean change was considered to be statistically significant.

RESULTS

Patients

There were no adverse reactions and all injections were well tolerated. N o change could be found

Journal of the American SocieD' of Echocardiograpby ZOtZ et al. 3 Volume 9 Number 1

. . . . . . . . . . n~ . . . . m-72-----7_ Z

F igu re 1 Example o f a contrast echocardiogram using intravenous Albuncx 0.12 m l / k g (A) at end diastole and (B) at end-systole.

4 Zotz et al. Journal of the American Society of Echocardiography January-February 1996

0,-----0

EDV H

o

ESV "_ ¢

E F o o

¢

-lbo-;o-;o-ao- o o 2'o io io go 1;o %

--o-- Observer 1

• Observer 2

Figure 2 Mean percent differences + their confidence intervals between native echocardio- graphic and angiographic volumes (end-diastolic [EDV] and end-systolic [ESV]) and ejection fractions (E.F). Values not containing zero are considered to be significantly different.

concerning pulse rate, respiratory rate, oxygen saturation, or temperature in any patient due to the injection of Albunex. There were no significant changes to preinjection values in any of the labortory parameters tested. No significant change could be found in the electrocardiograms due to the injection of Albunex.

Echocardiographic Results

In two thirds of all patients a good or sufficient left heart echocardiographic conta'ast was obtained. An example of contrast echocardiograms is demonstrated in Figure 1, which displays the end-diastolic (Figure 1, A) and end-systolic (Figure 1, B) echocardio- graphic pictures of a patient after the intravenous in- jection of echo contrast medium. The calculated mean differences to angiographic values from the tables show that both observers significantly underes- timated end-diastolic and end-systolic angiographic volumes (Figure 2) using native echocardiograms. However, the ejection fractions were not significantly different from angiographic ejection fractions. Con- versely, both observers overestimated angiographic volumes significantly, one observer by as much as 70% (Figure 3). Again, ejection fractions measured from contrast echocardiograms were not different from an- giographic ejection fractions concerning both observ- ers (Figure 3). Interobserver differences are displayed in Figure 4. In general, the confidencc intervals of the mean differences between observers were comparable for angiographic and contrast echocardiographic measurements. Concerning end-diastolic volumes, the observers did not differ significantly from each other when measuring angiographic and contrast echocardiographic volumes, whereas with respect to native contrast echocardiograms both observers dis- agreed. Concerning end-systolic volumes and the

ejection fraction, both observers disagreed when ana- lyzing angiographic and contrast echocardiographic pictures. The agreement between both observers on native echocardiograms seems most likely due to the great confidence interval of the mean differences be- tween both observers' measurements when analyzing native echocardiograms.

Endocardial Contour Finding

Ovcrall more segments were visualized with Albunex than without. Differences betwecn end-diastolic and end-systolic left heart opacification can be anticipated from Figurc 5. The relative proportion ofnot-visual- izcd segments after the injection of Albunex (higher dose) to not-visualized segments in native echocar- diograms was larger at end systole than at end diastole.

Safety

There were no changes in the laboratory variables measured or in the electrocardiograms. No patient complained of any scnsations or other sidc ef- fects.

DISCUSSION

Since the original observation of Gramiak and Shah 3 that the injection ofindocyanine green causes opaci- fication of blood flow, the need and possibility to visualize intracavitary and parenchymal blood flow 4 6

h a s been recognized and has stimulated worldwide research to develop biocompatible and efficacious contrast agents. 7-9 The value of contrast echocardi- ography to improve volume determination for the right ventricle has been proven. ~°'1~ Digital subtrac- tion of left ventricular contrast echocardiograms has already been shown to have significant limita-

Journal of the American Society of Echocardiography Volume 9 Number 1 ZOtZ et al. 5

EDV

ESV

o o - -o - - O b s e r v e r 1

• - - , • O b s e r v e r 2

O

E F o o

ID- - - - - - -4b

-1(10-;0 " 6 0 - 4 0 - 2 0 0 2; 410 6'0 8'0 1;0 %

Figure 3 Mean percent differences +__ their confidence intervals between contrast echocardio- graphic and angiographic volumes (end-diastolic [EDV] and end-systolic [ESV]) and ejection fractions (EF). Values not containing zero are considered to be significantly different.

E D V

E S V = =

D - -

EF

o 2'0 4'0 6'0 8'0 1;o %

A n g i o

. E c h o

• C o n t r a s t

Figure 4 Mean percent differcnces _+ their confidence intervals between measurements o1~ both observers for angiographic (Anglo), native (Echo) and contrast echocardiographic (Con- trast) volumes and ejection fractions. Values not containing zero are considered to be signifi.. cantly different.

dons, ~2"Ia and it was thus the aim of this study to evaluate the left heart contrast agent Albunex with respect to direct improvement of left ventricular en- docardial border delineation.

This study shows that Albunex reproducibly opaci- ties the left ventricle and can be used for left ventricu- lar endocardial border delineation leading to reduced interobserver variation concerning volume determi- nation. No clinically visible side effects were ob- served. There were also no changes of laboratory data or electrocardiograms.

The endocardial contours found using Mbunex in this study differed from those in the native echocar- diograms. An example is depicted in Figure 1. Volume determinations performed in the native echocardio- grams in this study were consistently lower than those obtained using contrast echocardiography (Figure 3).

This is due to the fact that the contrast agent flows into all the left ventricular trabeculations, delinearing the outer contour similar to that in cineventriculogra- phy. 1"2 Because of the filling of echo-deprived zones in the left ventricle by the contrast agent, which helped the observers in discriminating between contrast me- dium and tissue echoes, larger volumes were obtained using contrast echocardiography. However, the bright contrast information often interfered with the tissue information and thus resulted in volumes that were too large compared with angiographic volumes. Color superpositioning of contrast echocardiograms might represent an alternative to the fusion of con- trast and tissue information in bright contrast echocardiograms] ~ The further increase in volume compared with ventriculograms may be due to the point spread of the ultrasound targets by the cchocar-

6 Zotz et al. Journal of the American Society of Echocardiography J'anuary-February 1996

30

25

20

15

10

5

0

A

30

25

20

15

10

5

Endocardial border delineation at end-diastole using Albunex i.v.

number of segements not visualized

1 2 3 4 5

segment

Endocardial border delineation at end-systole using Albunex i.v.

number of segements not visualized

Non-Contrast I 2nd Injection

m Non-Contrast 1 F-/-/~2nd Injection ]

1 2 3 4 5

B segment

Figure 5 Endocardial border delineation at end diastole (A) and end systole (B) in five segments of the left ventricle. Segment 1 = lower septum; segment 2 = upper septum; segment 3 = apex; segment 4 = upper posterior wall; segment 5 = lower posterior wall.

diograph and may not be overcome by color superpo- sifioning or other measures of selective contrast visu- alization. The large difference between angiographic and contrast echocardiographic volumes raises the question whether " t rue" volumes might bc sought between angiographic and contrast echocardio- graphic volumes; this would need experimental justi- fication. In this study echocardiographic volumes were compared with angiographic volumes because angiography still represents the gold standard for volumetry, although there are principal methodologi- cal differences. Angiography represents a contour and radiographic method whereas echocardiography rep- resents a slice method. 1 It is conceivable, that more accurate volumes can be obtained with echocardio- graphic contrast agents as opposed to radiographic contrast agents when echocardiography is used. In this study a wide difference between observers when

analyzing conventional echocardiograms (Figure 4) could be reduced to a level that was comparable to that found when analyzing the corresponding angio- grams using Albunex. Because this study was per- formed in consecutive patients, about one third of paticnts showed inferior quality of native echocardio- grams that could be improved with Albunex.

Albunex is expected to be approved in the near fu- ture in several countries. Its efficacy and safety have been proven in numerous studies, 7as'16 but with a va- ricty of results and a high interobserver variability. 7 This variability could perhaps be attributed to differ- ent hemodynamic conditions in different patient groups. With a different agent (SHU 508A), a good correlation between pulmonary artery pressure and left heart opacification in patients with normal as well as with a wide range of abnormal prcssurcs ~7 could be demonstrated. Because this could have been related

Journal of the American Society of Echocardiography Volume 9 Number 1 Z o t z et al. 7

to cardiac output as well, two studies could relate the transit time of contrast material to cardiac o u t p u t , 18"19

realizing the need for a stable ultrasound scatterer.

Advantage of Myocardial Contrast Echocardiography

Studics of Sabia et al.20 and Ito et al. 2~ have shown that myocardial contrast echocardiography can pro- vide prognostic information after acute myocardial infarction and thrombolysis. It also offers a possibil- ity of studying viable myocardium per fused by col- laterals at the time of cardiac catheterization 22-24 and coronary flow reserve. In all these studies the con- trast agents were radiographic agents produced by the authors themselves. The availability of an agent that did not change pulmonary or systemic hemody- namics and had no influencc on ejection fraction or wall motion would be an advantage and progress.

It has been shown that this standardized agent can be used in patients without increasing pulmonary artery pressure) s The lack of efficacy in some pa- tients in this study, probably relating to systolic left ventricular pressure, maizes it necessary to augment the stability of the shell. However, the properties of Albunex mal<e it a valuable agent for left ventricular function studies.

Criticism of the Method

A systolic reduction of contrast information can be anticipated from Figures 1 and S. It is possible that this reduccd opacification is caused by the systolic pressure in thc left vcntricular cavity causing the mi- crobubbles to collapse. Howcver, wc could not find any link bctwcen catheterization data and echo per- formance of Albunex, although this study was not intended to do so. It is likely that the limited left vcn- tricular opacification during systole in part caused the observed differences to angiographic volumcs. It is thus conceivable that with an incrcased stability of the agent (i.e., thc ultrasound scattcring microbubbles during systole), the precision and reproducibility of the mcthod can be improved.

It is strildng that the underestimation of cnd- diastolic volumes using convcntional echocardiogra- phy in this study is higher than, for example, in the study by Himelman et al. 26 While hc dcscribes a mean underestimation for end-diastolic volumes of 11%, this value was twice as high as the correspond- ing value in this study. In this study, however, we did not average several heart cycles to minimize biologi- cal variability as in the cited study. Conversely, end- systolic values in this study werc not undcrestimatcd to the degree they were in the study by Himelman et al. To our knowledge no data have been published that analyze the relationship betwcen contrast and

angiographic volumes. Although angiographic vol- umes correlate well to pathologic specimens, it can only be speculated whether contrast echocardio- graphic volumes correlate better to excised and pre- pared left ventricles than angiographic volume deter- minations in these ventricles. 2

Possibly, as this study indicates, the use of contrast material can contribute to a reduction of the interob- server variability reported in other studies .27,28 An in- teresting finding of this study is the fact that although there are marked deviations of volume measurements, the ejection fraction of the three methods tested are very similar. This suggests that the overestimations and underestimations are proportional for both dias- tole and systole, and thus will not affect the ejection fraction measurement. This has been neted previously by Morrissey et a l . 29 If this observation l~olds true, the potential benefit of contrast for enhanced border de- tection may be a mute point for the assessment of ventricular function. However, since Figure 5 indi- cates a systolic loss of contrast and th~as a decreased end-systolic border delineation, these J:esults need to be reproduced with a pressure-stable agent. Also, the improved interobserver variability may imply a poten- tial improvement of volume determination as well.

The way volume determination using automatic border detection algorithms (using for example the integrated backscatter signal) is influenced by the use of contrast 29 needs to be determined. 3dso, the use of color superpositioning remains an alternative, and its value for clinical routine 3° after intravenous use of echo contrast agents needs to be inw:stigated. With both alternatives, a reduction in the echocardio- graphic volume underestimation and contrast echo- cardiographic overestimation of volumes may be- come a reality.

Inasmuch as lung-passing echo contrast agents are now available, improved volume determination using contrast echocardiography is likely to become a clini- cal and noninvasive reality. However, our experience with SHU 508 and Albunex has shown that left heart opacification is dependent on pulmonary artery pres- sure for SHU S08A 17 and possibly on left ventricular pressure for Albunex as this study poasibly indicates.

There still remain 10% to 15% of patients in whom one cannot obtain satisfactory transthoracic echocar- diograms because of emphysema, adipositas, and thorax malformations. In these patients the trans- esophageal echocardiographic approa~ch would seem to be the appropriate alternative.

Clinical Implication

Since it has bcen dcmonstratcd in other studies as well as in this study that contrast cchocardiography is a safe procedure, '5 this modality should be consid-

Journal of the American Sociew of Echocardiography 8 Zotz et al. lanuary-February 1996

e red a ser ious a l te rna t ive to c o n v e n t i o n a l c inevcn-

t r i cu lography , especial ly in pa t i en t s w i t h an al lergy to

c o n v e n t i o n a l con t ras t agen ts and those w i t h p o o r lef t

ven t r i cu la r func t ion , h y p e r t h y r o i d i s m , and i m p a i r e d

renal f u n c t i o n o b v i a t i n g the n e e d for c a the t e r i z a t i on

o f t he lef t vent r ic le .

REFERENCES 1. Erbel R, Schweizer P, Lambertz H, et al. Echoven-

triculography: a simultaneous analysis of two-dimensional echocardiography and cineventriculography. Circulation 1983;67:205-15.

2. Erbel R, Krebs W, Henn G, et al. Comparison of single-plane and biplane volume determination by two-dimensional echo- cardiography. I. Asymmetric model hearts. Eur Heart J 1982;3:469-80.

3. Gramiak R, Shah PM. Echocardiography of the aortic root. Invest Radiol 1968;3:356-8.

4. Meltzer R, Serruys P, McGhie ], Verbaan N, Roelandt J. Pulmonary wedge injections yielding left-sided echocardio- graphic contrast. Br Heart J 1980;44:390-6.

5. Feinstein SB, Ten Cate FJ, Zwehl W, et al. Two-dimensional contrast echocardiography. I. In vitro development and quan- titative analysis of echo contrast agents. J Am Coil Cardiol 1984;3:14-20.

6. Ten Cate F, Feinstein S, Zwehl W, et al. Two-dimensional contrast-echocardiography. II. Transpulmonary studies, l Am Coil Cardiol 1984;3:1035-42.

7. Feinstein SB, Cheirifl[, Ten Cate FJ, et al. Safety and efficacy of a new transpulmonary ultrasound contrast agent: initial multi- center clinical results. I Am Coil Cardio11990;16:316-24.

8. Berwing K, Schlepper M. Echocardiographic imaging of the left ventricle by peripheral intravenous injection of echo con- trast agent in patients. Am Heart ] 1988;115:399-408.

9. Zotz R, Erbel R, Walch A, Krone V. A new biocompatible and biodegradable contrast agent for left heart opacification [Ab- stract]. Circulation 1991;84-II:161.

10. Wann LS, Stickels KS, Bamrah VS, Cross CM. Digital pro- cessing of contrast echocardiograms: a new technique for measuring right ventricular ejection fraction. Am J Cardiol 1984;53:1164-8.

11. Royal D, Nissen S, Elion I, Distante A, DeMaria A. Limita- tions of digital subtraction contrast echocardiography in en- hancing left ventticular endocardial definition. Am Heart J 1987;113:1437-44.

12. Grube E, Becher H, Backs B. Automatic contour finding and digital subtraction technique in 2-Dimensional echocardiog- raphy. In: Roelandt I, ed. Digital techniques in echocardiogo raphy. Dordrecht: Martinus Nijhoff Publishers, 1987.

13. Ten Cate FJ. Myocardial contrast 2-Dimensional Echocar- diography: Analysis of myocardial perfusion. In: Roelandt J, ed. Digital techniques in echocardiography. Dordrecht: Mar- tinus Nijhoff Publishers, 1987:39-46.

14. Clas W, Brennecke R, Zotz R, ErbeI R, Jung D, Meyer J, Colour superposition: a new modality for contrast echocardi- ography. Int J Card Imaging 1987;2:111-6.

15. Keller MW, Glasheen W, Kaul S. Albunex: a safe and effective

commercially produced agent for myocardial contrast echo- cardiography. J -&,a Soc ECHOCae, DIOGR 1989;2:4842.

16. Ten Cate FJ, Widimsky P, Cornel JH, Waldstein DJ, Serruys PW, Waaler A. Intracoronary Albunex. Its effects on left ventricular hemodynamics, function and coronary sinus flow in humans. Circulation 1993;88:2123-7.

17. Zotz RJ, Genth S, Erbel R, Meyer J. Contrast echocardiogra- phy of the left ventricle--an independent predictor of pulmo- nary artery pressure? Int J Card Imaging 1994;10:195-203.

18. Galanti G, Jayawera AR, Villaneuva FS, Glasheen WP, Ismail S, Kaul S. Transpulmonary transit of micmbubbles during contrast echocardiography: implications for estimating cardiac output and pulmonary blood volume. J AM Soc ECHOCaRDIOGR 1993;6:272-8.

19. Zotz R, Genth S, Krone V. Correlation of thermodilution cardiac output and indicator dilution curves from contrast echocardiograms with biodegradable microparticles [Ab- stract]. Circulation 1993;88-I:163.

20. Sabia PJ, Powers ER, Jayawera AR, Ragosta M, Kaul S. Functional significance of collateral blood flow in patients with recent myocardial infarction. Circulation 1992;85: 2080-9.

21. Ito H, Tomooka T, Sakai N, et al. Lack of myocardial perfu- sion immediately after successful thrombolysis: a predictor of poor recovery of left ventricular function in anterior myocar- dial infarction. Circulation 1992;85:1699-705.

22. Widimsky P, Cornel JH, Ten Cate FJ. Evaluation of collateral blood flow by myocardial contrast enhanced echocardiogra- phy. Br Heart J 1988;59:20-2.

23. Grill HI', Brinker IA, Taube JC, et al. Contrast echocardio- graphic mapping of collateralized myocardium in humans before and after coronary angioplasty. J Am Coil Cardiol i990;16:1594-600.

24. Porter TR, D'Sa A, Turner C, et al. Myocardial contrast echocardiography for the assessment of coronary blood flow reserve: validation in humans. J Am Coil Cardiol 1993;21: 349-55.

25. Zotz RI, Genth S, Mitrovic V, Waaler A, Meyer J. The influence of Aibunex on the puhnonary circulation in patients with pulmonary hypertension or left heart failure. (in press).

26. Himelman RB, Cassidy MM, Landzberg JS, Schiller NB. Reproducibility of quantitative two-dimensional echocardi- ography. Am Heart J 1988;115:425-31.

27. Gordon EP, Schnittger I, Fitzgerald PJ, Williams P, Popp RL. Reproducibility of left ventricular volumes by two-dimen- sional echocardiography. J Am Coil Cardiol 1983;2:506-13.

28. Schnittger I, Fitzgerald PJ, Daughters GT, et al. Limitations of comparing left ventricular volumes by two-dimensional echocardiography, myocardial markers and cineangiography. Am J Cardiol 1982;50:512-9.

29. Morrissey RL, Siu SC, Guerrero L, Newell IB, Weyman AE, Picard MH. Automated assessment ofventricular volume and function by echocardiography: validation of automated bor- der detection. J AM Soc ECHOCARDIOGR 1994;7:107-15.

30. Zotz R, Genth S, Rupprecht HI, Brennecke R, Erbel R, Meyer J. Left ventricular volume determination using colour superpositioning of contrast echocardiograms. Eur J Ultra- sound 1995;2:17-27.