Peripheral venous contrast echocardiography

METHODS

Peripheral Venous Contrast Echocardiography

JAMES B. SEWARD, MD ABDUL J. TAJIK, MD, FACC DONALD J. HAGLER, MD, FACC DONALD G. RITTER, MD, FACC Rochester, Minnesota

Contrast echocardiography is the technique of injecting various echoproducing agents into the bloodstream and, with standard echocardiographic techniques, observing the blood flow patterns as revealed by the resulting cloud of echoes. These techniques have only recently been utilized to evaluate various cardiac defects. Two physical properties of these agents characterize their usefulness: (1) clouds of echoes can be observed downstream as well as at the injection site, and (2) the echoproducing quality of these agents is completely lost with a single transit through either the pulmonary or the systemic capillary bed. Thus, detection of resultant echoes in both the venous and the arterial blood pool is indicative of abnormal shunting. In 60 patients with a spectrum of cardiac defects and a wide range in age of presentation, studies were made of (1) the feasibility of performing contrast echocardiography with superficial peripheral venous injections, and (2) the clinical usefulness of this relatively noninvasive technique in detecting and localizing intracardiac right to left shunting. Most superficial peripheral veins could be utilized, and the resultant contrast echograms were reproducible and similar in quality to those obtained with more central (caval) injections. Right to left shunts could be localized to the atrial, ventricular or intrapulmonary level. Characteristic flow patterns were also recognized for tricuspid atresia and common ventricle.

T h e concept of c o n t r a s t e c h o c a r d i o g r a p h y was i n t r o d u c e d in the m i d 1960's, b u t the t e c h n i q u e was utilized only sporadically a n d p r i m a r i l y to verify cardiac a n a t o m y as assessed with ultrasound. 1-3 R e c e n t l y use of the method has been extended to the evaluation of blood flow patterns in p a t i e n t s with various congenital h e a r t defects. 4 P r e v i o u s studies utilizing central venous injections of e c h o - p r o d u c i n g agents have established this technique as a sensitive m e a n s of detecting i n t r a c a r d i a c right to left shunts. 4,s T h e p u r p o s e of this s t u d y was, first, to evaluate the feasibility of p e r f o r m i n g c o n t r a s t e c h o c a r d i o g r a p h y with use of superficial p e r i p h e r a l venous injections and, second, to d e t e r m i n e the clinical usefulness of this relatively noninvasive t e c h n i q u e in detecting and localizing i n t r a c a r d i a c right to left shunts. Methods

From the Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Manuscript received July 12, 1976, accepted August3, 1976. Address for reprints: James B. Seward, MD, Mayo Clinic, Rochester, Minnesota 55901.

202

Standard M mode echocardiograms were obtained with a Smith-Kline instrument (Ekoline 20 ultrasonoscope) interfaced with a Cambridge recorder. All patients were examined in the recumbent position by means of already described echocardiographic techniques. 6 For most examinations we used a 2.25 megahertz, 7.5 or 5 cm focused transducer and in infants a 5.0 megahertz nonfocused pediatric transducer. Sixty patients with various congenital and acquired cardiac defects were studied (Table I). Their age ranged from 2 months to 65 years (mean 20 years), their weight from 3.9 to 87 kg and their height from 45 to 185 cm. Technique: All patients were studied during cardiac catheterization, when an intravenous route was routinely established in a superficial vein (hand, 57

February1977 The American Journal of CARDIOLOGY Volume39

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SEWARD ET AL.

patients; antecubital, 2; foot, 1). In adult patients a 16 or 18 gauge (5 cm) Teflon ®venous sheath was utilized. In children, a 20 gauge over-the-needle (3.8 cm) Teflon venous sheath or a 23 gauge (1.9 cm) thin-walled vein needle was used. A double-syringe technique (Fig. 1), with injection of 1 ml of indocyanine green dye followed by a manual flush of 5 to 10 ml of isotonic saline solution into vein, was used to produce the contrast effect. An echocardiogram was recorded during each injection. The concentration of dye was varied according to body weight: 5 mg/ml for patients weighing more than 30 kg, 2.5 mg/ml for those 10 to 30 kg, 1.25 mg/ml for those 5 to 10 kg, and 0.625 mg/ml for those less than 5 kg. An average of two peripheral venous injections were performed in each patient to determine reproducibility. In most patients, central venous injections of indocyanine green dye were also per-

formed during the cardiac catheterization procedure, and the results were used for comparison. In this presentation two standard transducer positions (comparable to positions 3 and 4 of Feigenbaum 7) were uti, lized to record and characterize the echocardiographic contrast blood flow patterns. In transducer position 3 (Fig. 2, left), with the transducer in the third or fourth intercostal space along the left sternal border and directed posteriorly, the right ventricular cavity, ventricular septum, left ventricular cavity and mitral valve echoes were recorded. Special care was taken so that the area encompassed by the mitral valve echo (mitral valve orifice or funnel) could be clearly visualized. In transducer position 4 (Fig. 2, right), with the transducer directed superiorly and medially, the right ventricular outflow tract, aortic root and left atrium were recorded.

Observations and Results After p e r i p h e r a l venous injections, the r e s u l t a n t c o n t r a s t echograms were reproducible a n d similar in quality to those o b t a i n e d with m o r e central injections. Occasionally the c o n t r a s t effect was less dense with p e r i p h e r a l injections, b u t use of a slower p a p e r speed helped to accentuate the contrast density. T h e injection of indocyanine green dye with saline solution was consistently superior to injection of either saline solution or blood alone in producing a dense cloud of echoes (Fig. 3). However, m a n y substances when used alone (blood, saline solution, indocyanine green dye) p r o d u c e d satisfactory contrast, particularly in infants and in patients with polycythemia, severe tricuspid or p u l m o n a r y regurgitation or low o u t p u t state.

FIGURE 1. Simulated demonstration of the double syringe technique, injecting into a superficial hand vein. Injection of 1 ml of indocyanine green (left) followed by a manual flush of 5 to 10 ml of isotonic saline solution (right) was used to produce the echographic contrast ef-

TABLE I

Congenital and Acquired Cardiac Defects Studied

fect.

Patients (no.)

Cardiac Defect

Position (3)

I

Position (4)

Hl HllIH ll HIH Hl,IHHllll Hl't lll!l Llt NLIl lN!t

.~,~.~

~

".,~,

•

'

"'

k,.:'::"

"

FIGURE 2. Echocardiographic scan from the mitral valve level to the aortic root (left to right). Left, position 3: right ventricle (RV), ventricular septum (VS), mitral valve (MV) and mitral valve orifice or funnel (MVF) are recorded with the transducer along the left sternal border and directed posteriorly. Right, position 4: right ventricular outflow (RVO), aorta, aortic valve (AV) and left atrium (LA) are recorded with the transducer directed superiorly and medially.

Group I No right to left shunt (19 patients) Normal Cardiomyopathy Congestive Obstructive Aortic stenosis Aortic insufficiency Pulmonary stenosis Primary pulmonary hypertension Atrial septal defect Ventricular septal defect Postoperative Aortic stenosis Tetralogy of Fallot Group II Right to left shunt, atrial level (11 patients) Ostium secundum, isolated Sinus venosus defect Patent foramen ovale Atrioventricular canal Partial Complete Right to left shunt, ventricular level (24 patients) Ventricular septal defect Isolated Tetralogy of Fallot Pulmonary atresia Truncus arteriosus Intrapulmonary right to left shunt (1 patient) Osler-Weber-Rendu Group Ill (5 patients) Tricuspid atresia Common ventricle

February 1977

The American Journal of CARDIOLOGY

Volume 39

3 3 2 1 2 1 1 1 3 3 2 1 1 4 1 t 5 3 2 1 5 14 4 1 2 3

203

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SEWARD ET AL.

,~,-,roar

•

,v-~- 'rTr " [ g ~"

r ' f f T'TT q'"-V J'[,/ \ t - ~ r;~r T '

MV~

' '

,

'

' '

~

"

B ! '~F'I '1II!!IF I! i~,!J',, 1:1' I 'IT i~!~t'i~!'t'~l']lll'Jl]l!?~l ~!r]P'qP l' i!il!!;!r,l!!i!l~,!;il![i!ti!il!!ill!lli]!!itl;i';l!'iilii!ll~:ll Ii~1!!:%i;l!illHlilill!!!i'i

FIGURE 3. Contrast echocardiograms from a 27 year old man after repair of tetralogy of Fallot. There were no residual shunts. Both echograms were performed sequentially. A, a dense cloud of echoes appeared during ventricular diastole in the right ventricle after injection of indocyanine green dye into a superficial hand vein. B, without change in echographic signal a similar injection was performed with 10 ml of isotonic saline solution. Note the superior contrast effect after injection of indocyanine green dye compared with the effect after injection of saline solution. MV = mitral valve; RV = right ventricle; VS = ventricular septum.

In the majority of patients an acceptable contrast effect could be obtained. Difficulties in demonstrating clouds of echoes occurred primarily in obese or emphysematous subjects, in whom a good quality control standard M mode echogram could not be obtained. Approximately 5 percent of contrast injections were discarded because of poor diagnostic quality and were not included in this presentation: The results of our study are based on a previously established observation that, after a systemic venous injection of indocyanine green dye, the resultant echoes remain confined to the right heart chambers and normally do not appear in the left heart chambers. Because the contrast effect of these echo-producing agents is usually completely lost with a single transit through the pulmonary capillary bed, appearance of any echoes in the left side of the heart after a systemic venous injection signifies the presence of an intracardiac or intra-

204

February 1977

The American Journal of CARDIOLOGY

pulmonary right to left shunt. On this basis, the 60 patients were classified into three groups according to the observed contrast flow patterns (Table I): group I, 19 patients who had no evidence of a right to left shunt; group II, 36 patients who had evidence of a right to left shunt at the atrial (11 patients), ventricular (24 patients) or intrapulmonary (1 patient) level; and group III, 5 patients with a specific flow pattern of tricuspid atresia (2 patients) or common ventricle (3 patients). Group I

No patient in group I had a right to left shunt demonstrated with the standard indicator-dilution curves. After a peripheral venous injection of indocyanine green dye, the contrast effect appeared and remained confined to the right heart chambers (Fig. 4). Contrast echoes delineated boundaries of the right ventricular cavity and

Volume 39

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SEWARD ET AL.

MV~

B

I

v~,

I j

.",.r~.":.',' .' ~' ' .'., ' ,,..~'~','.... \'

~ i

!

~=1

I,Iii~!1 ¸ ~E I!1~1~'1

~

~ r !

I

!

r

L

FIGURE 4. Contrast echocardiograms. A, from a 23 year old woman with pulmonary stenosis and no demonstrable shunt. After antecubital venous injection of dye, a cloud of echoes appeared during ventricular diastole and remained in the right ventricle for approximately 7 cardiac cycles. No contrast medium appeared in the left ventricle. MV = mitral valve; RV = right ventricle; VS = ventricular septum. B, from a 5 year old boy with mild congenital aortic stenosis and no intraoardiac shunt. After superficial hand venous injection of dye, a cloud of echoes appeared during diastole and remained confined to right ventrioular outflow (RVO). AV = aortic valve; LA = left atrium.

LA m

outflow tract and outlined the right side of the ventricular septum and atrial septum. This contrast effect normally persisted for four to six cardiac cycles but in certain circumstances (that is, in the presence of severe valve regurgitation or low output state), it persisted longer (Fig. 5). The three patients with an isolated secundum atrial septal defect and three with an isolated ventricular septal defect had a predominant left to right shunt and no evidence of an intracardiac right to left shunt with standard caval indicator-dilution dye curves performed during cardiac catheterization. Contrast echocardiography also failed to detect a right to left shunt in these six patients.

Group II In all 36 patients in this group, after peripheral venous injection the cloud of echoes not only was noted in

~ .

J~P.,

i

".

"

". . . . . . . .

'..

",

",'~,'m'r.i'-,m.~a~-T~

~ : .

"~

"

;~ff.

the right heart chambers but also appeared in the left heart chambers, indicating the presence of a right to left shunt. In this group the recognition of three distinct flow patterns allowed localization of the site of the right to left shunt to the atrial (11 patients), ventricular (24 patients) or intrapulmonary (1 patient) level. R i g h t to l e f t s h u n t at a t r i a l level: In 11 patients a right to left shunt was demonstrated with dye-dilution curves at cardiac catheterization (range 5 to 50 percent), and in all of these patients peripheral venous contrast echocardiography demonstrated a right to left shunt at the atrial level (Table I). With transducer position 3 (viewing the mitral valve funnel), the flow pattern of right to left shunting at the atrial level was characterized by the appearance of t h e cloud of echoes in the right ventricle during ventricular diastole and the simultaneous appearance of a similar

February 1977

The American Journal of CARDIOLOGY

Volume 39

205

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SEWARD ET AL.

LV

FIGURE 5. Contrast echocardiogram from a 14 year old boy with congestive cardiomyopathy. After superficial

hand venous injection of dye, echoes appeared in the right ventricle (RV) during diastole. Consistent with a low output state, contrast echoes persisted in the right ventricular cavity for more than 20 cardiac cycles (complete record not shown). LV = left ventricle; MV = mitral valve; VS = ventricular septum.

cloud of echoes in the left ventricle through the mitral valve funnel (Fig. 6A). As previously reported, echoes appearing in the left ventricle may occasionally be delayed by one and rarely by two cardiac cycles.* After complete mitral valve opening, echoes subsequently appeared in the left ventricular outflow tract anterior to the mitral valve echo. With transducer position 4 (viewing the right ventricular outflow tract, aortic root and left atrium), the cloud of echoes appeared first in the left atrium, usually during ventricular systole; second in the right ventricular outflow tract during late diastole; and third in the aortic root during subsequent systole (Fig. 6B). However, in 5 of the 11 patients, contrast echoes in the left atrium were either delayed in appearance or were not visualized at all in spite of right to left shunting at the atrial level (Fig. 6C). This situation is believed to be a result of the streaming of dye low in the left atrium while the echocardiographic beam is traversing the mid to upper part of the left atrium. Because of this potential diagnostic problem, we have found transducer position 3 (viewing the mitral valve funnel) most sensitive for demonstrating the characteristic flow pattern of a right to left shunt at the atrial level (that is, the initial appearance of echoes in the mitral valve funnel). R i g h t to left s h u n t at the v e n t r i c u l a r level: With contrast echocardiography we were able to demonstrate a right to left shunt at the ventricular level in 24 patients. All showed a right to left shunt with standard dye-dilution curves at the time of cardiac catheterization. One patient had an isolated ventricular septal defect with severe pulmonary hypertension (pulmonary resistance 40.0 units/m2). The remaining 23 patients of this group had tetralogy of Fallot (5 patients), pulmonary atresia with ventricular septal defect (14 patients) or truncus arteriosus (4 patients). With transducer position 3 (viewing the mitral valve funnel), after a peripheral venous injection of dye the flow pattern of a right to left shunt at the ventricular

206

February 1977

The American Journal of CARDIOLOGY

level was characterized by the appearance of the cloud of echoe~ first in the right ventricle during ventricular diastole (Fig. 7A). During subsequent systole, no echoes were seen in the left ventricle. However, echoes started to appear in the left ventricle early in diastole of the following cardiac cycle. Further analysis of the timing of this right to left shunt revealed that the shunt flow commenced during the isovolumic relaxation phase of diastole (that is, after aortic valve closure and before mitral valve opening) (Fig. 7A). In these patients echoes in the left ventricle appeared and remained anterior to the mitral valve, with the mitral valve funnel remaining essentially echo-free (undyed blood through the mitral valve). With transducer position 4 (viewing the right ventricular outflow tract, aortic root and left atrium), the cloud of echoes appeared first in the right ventricular outflow tract during mid to late ventricular diastole and then appeared in the aortic root during subsequent systole, with the left atrium remaining echo-free (Fig. 7B). As discussed, a similar flow pattern (late diastolic appearance in the right ventricular outflow tract and systolic appearance in the aortic root) may be observed in certain patients with a right to left shunt at the atrial level when a similar transducer position is used. For this reason we believe it is important to observe the mitral valve funnel in order to distinguish a right to left shunt at the ventricular level from a similar shunt at the atrial level. Evidence of a right to left shunt was observed in all patients with transducer position 4 (systolic appearance of echoes in aortic root) and in 22 of 24 with transducer position 3 (diastolic appearance of echoes in the left ventricular outflow) (Fig. 7C). No patient had aortic valve insufficiency. We therefore recommend that to detect a right to left shunt, both transducer positions should be viewed. However, transducer position 3 (viewing mitral valve funnel) is most helpful in localizing such a shunt.

Volume 39

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SEWARD El" AL.

C ."'~'!:: :-,~..?~:":~."~:"~".' : •

FIGURE 6. Contrast echocardiograms. A, from a 60 year old woman with a sinus venosus atrial septal defect (pulmonary to systemic flow ratio 2.24). After peripheral hand venous injection of dye, echoes appear during diastole in the left ventricle through the mitral valve funnel (MVF) ( a r r o w h e a d ) nearly simultaneously with the appearance of echoes in the right ventricle. MV = mitral valve; r v = tricuspid valve; VS = ventrioular septum. B, from a 12 year old girl with a complete A-V canal. After hand venous injection of dye, dye contrast initially appears in the left atrium (LA) (1), then in the right ventricular outflow area (RVO) during diastole (2) and, with the next systole, in the aortic root (Ao Root) (3). C, from a 1 year old infant boy with complete A-V canal. After hand venous injection of dye, a cloud of echoes appears in the right ventricular outflow area (RVO) during ventricular diastole and in the aorta with subsequent systole. Only a small contrast effect is recorded in the left atrium (LA). AV = aortic valve.

-

.

.

.

.•

,

."

..o,

,..-

.

~_..~- ~

__

~

.

..,

~

,:

-

-

:..

.

i

--

.

1

,

~v

.

.

.

.

I

' %,

~ ;t

~

,

, ~,, ~

_rE .~'~_""'_'-. • ~ " ~ ' , - ' ~ ,~:.~.

February

1977

,

'" "

!:,

~.

!,~

~ :' _

.

,,~~

, ,' ,~ '

The American

',

~

"

'

,

__ ~ - - _

~

r

.

;

.

,~,,,~ '

.

. -'-'~. . . . . . . . .

J o u r n a l of C A R D I O L O G Y

'k

. . . .

J~', . . . . .

Volume

39

"vd

'

~,,~-...~,.

207

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SEWARD ET AL.

t,1, . . .~ .

'

,r,,w

~

'

''::'tii:li'llLi!,

-,,-~r,- .'wm', ~

,

.. . . .

~

~

T ..,/

~,,-.,.-,

~i;i:i,~;

February

,,,

-

~..... ~.. , .

~

=

1977

~---,~

~

~

American

" '

',~

,

~ -

. ,~,;,.~'

Journal

~ ,i

~

~

'

'

The

i:ii~

~

.~_ _ ~ . ~ -

,

- -' ~ .

:!i;izi~ii',!

:,

208

'prv~, "t'~,~ w-'~, "~,~

~

~,~,~,~_.,.~

• .-..,

. • . - - ~~ . , ~ ~

,,I~......:!! :t~,!~t'

• ~:~ . ' " ~ ' , . ~ ' ~ Z , 1 ~ " ~ , ~

~.~,~,~..~,,,~,~,

~

'~'F"

r

i;l~:t,;=!!ili!:ii'iii

.

'~'T,~

~

• ' : ~ ~v n..;

Ciii

I:'~.i~~ll!li~l!il;l~!!!l!;ltl~i"li!

,

:

.

~

~,z

:

i :

....

:

,,. , . ~ . ~ . ~

i;

.,,,,

-_._-=___

_

'

~

~¢-.~,

.

,-.~-- " , ~

.,.;..

:..

,.

of CARDIOLOGY

Volume

39

FIGURE 7. Contrast echocardiograms. A, from a 5 year old girl with pulmonary atresia and ventricular septal defect. After hand venous injection, a cloud of echoes initially appeared in the right ventricle (RV) during diastole. No echoes appeared in the left ventricle during subsequent systole. With the next diastole, echoes first appeared in the left ventricle anterior to the mitral valve echo (MV). Note that dye into the left ventricle begins to appear just before mitral valve opening (perpendicular) with the mitral valve funnel (MVF) remaining echo-tree. VS = ventricular septum. B, from a 41 year old woman with tetralogy of Fallot. After hand venous injection, dye initially appears during late diastole in the right ventricular outflow (RVO) area and with subsequent systole in the large aortic root (Ao). The small left atrium (I_A) remains echo-free. AV = aortic valve. C, from a 9 year old girl with pulmonary atresia and a ventricular septal defect. Echographic scan from transducer position 3 (mitral valve) to transducer position 4 (aortic root) is recorded during peripheral venous injection of indocyanine green dye. At the mitral level there is no detectable right to left shunt; however, with a scan to the base of the heart a dense cloud of echoes fills the aortic root (Ao Root). Ao = aorta; LA = left atrium; MV = mitral valve; RV = right ventricle; VS = ventricular septum.

PERIPHERAL VENOUS CONTRAST ECHOCARDIOC-~APHY--SEWARD ET AL.

FIGURE 8. A 25 year old woman with Osler-Weber-Rendu disease and diffuse pulmonary arteriovenous fistula. After hand venous injection, resultant echoes first appear during diastole in right ventricle (RV) and subsequently, with a delay of approxirr~ately four cardiac cycles, appear in the left ventricle from behind the mitral valve echo (MV). MVF = mitral valve funnel; VS = ventricular septum. The recording at left is made at a slower paper speed than that at right.

FIGURE 9. Contrast echocardiograms. A, from a 21 year old man with tricuspid atresia and dextrotransposed great arteries. With the transducer oriented to visualize the mitral valve (MV) and the small anteriorly located right ventricular chamber (RV), the resultant echoes, after hand venous injection of dye, initially appear during diastole in the left ventricle from behind the mitral valve echo (arrow). With subsequent systole, echoes appear in the right ventricular chamber (arrowheads). MVF = mitral valve funnel; VS = ventricular septum. B, from a 12 year old boy with tricuspid atresia and dextrotransposed great arteries. With the transducer in position 4, viewing the posterior great artery (pulmonary artery) and left atrium (LA) after hand venous injection of dye, echoes initially appear in the left atrium and subsequently appear during early systole in the pulmonary artery and in late systole in the right ventricular outflow tract (RVO). PV = pulmonary valve.

February 1977

The American Journal of CARDIOLOGY

Volume 39

209

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SicWARD ET AL.

Intrapulmonary location: An intrapulmonary right to left shunt was demonstrated in one patient (Table I) who had a diffuse pulmonary arteriovenous fistula associated with Osler-Weber-Rendu disease, s As noted, normally the echographic contrast effect is totally lost with a single transit through the pulmonary capillary bed. 2 However, in the patients with a pulmonary arteriovenous fistula, after injection of indocyanine green dye into the pulmonary trunk, echoes were detected in the left heart chambers, indicating presence of an intrapulmonary right to left shunt, s In the patient with a diffuse pulmonary arteriovenous fistula, with peripheral venous injections of indocyanine green and viewing the mitral valve (transducer position 3), a cloud of echoes initially appeared in the right ventricular chamber, and subsequently after an unusual delay (four to six cardiac cycles) these echoes appeared in the left ventricular chamber through the mitral valve funnel (Fig. 8). With transducer position 4 (viewing the right ventricular outflow tract, aortic root and left atrium), the cloud of echoes first appeared in the right ventricular outflow tract and then after a delay of four to six cardiac cycles appeared in the left atrium and aortic root. The one characteristic feature of right to left shunting secondary to one or more pulmonary arteriovenous fistulas is the prolonged delay before the appearance of contrast echoes in the left heart chambers after their initial appearance in the right heart chambers. Group III In addition to the localization of right to left shunt, certain contrast echocardiographic blood flow patterns were an aid in identifying specific complex congenital cardiac defects. Tricuspid atresia: The echocardiographic features of tricuspid atresia with a ventricular septal defect include a single demonstrable atrioventricular (A-V) valve and small right ventricular chamber. 9 Characteristic contrast echocardiographic flow patterns have been recognized in the presence of tricuspid atresia. 4 After a peripheral venous injection of dye and the transducer oriented to view the A-V valve and valve funnel (transducer position 3), the resultant cloud of echoes first appeared in the left ventricle through the mitral valve funnel during ventricular diastole (Fig. 9A). Usually with the subsequent systole the cloud of echoes appeared in the hypoplastic right ventricular chamber. This blood flow pattern has to date been most suggestive of tricuspid atresia. In transducer position 4 (viewing the right ventricular outflow tract, great artery and left atrium), the cloud of echoes appeared initially in the left atrium during ventricular systole and then in the posterior great artery and right ventricular outflow tract with subsequent ventricular systole (Fig. 9B). The contrast echoes entering the right ventricular outflow tract were consistently observed in mid to late ventricular systole. The appearance of echoes in systole as opposed to diastole in the right ventricular outflow tract distinguishes this flow pattern from that of right to left shunt 210

February 1977 The American Journal of CARDIOLOGY

associated with isolated atrial or ventricular septal defect. Common ventricle: The characteristic echocardiographic features of common ventricle with two A-V valves (a ventricular chamber that receives blood from two separate A-V valves) 1°-12 include: (1) the ability to record simultaneously two A-V valves without intervening septum, and (2) absence on an apex to base scan of a normally positioned ventricular septum separating the two A-V valves. 13 In patients with common (or single) ventricle with two A-V valves, a very characteristic echographic contrast flow pattern was observed. With two simultaneously recorded A-V valves, after a peripheral venous injection of dye, echoes initially appeared in the tricuspid valve funnel and in the same diastolic cycle appeared anterior to the mitral valve echo (Fig. 10A). Further analysis revealed that the echoes anterior to the mitral valve appeared after complete opening of the mitral valve (that is, after the E point) and not during the isovolumic relaxation phase of diastole as typically seen in patients with a right to left shunt at the ventricular level (see preceding). Even when two A-V valves cannot be visualized simultaneously, a resultant echographic contrast flow pattern showing arrival of the cloud of echoes anterior to the mitral valve during the rapid inflow period of ventricular diastole (after the mitral valve E point) is the strongest indirect evidence of common ventricle with two A-V valves (Fig. 10B). As we have described, a right to left shunt through a large ventricular septal defect is distinctly different from that observed in common ventricle. Caution: With sudden changes in acoustic impedance produced by the resultant contrast effect, an "overload" phenomenon may occur which appears as a direct posterior extension of the most dense contrast echoes (Fig. 11). Because these echo reverberations may extend from anterior into posterior chambers, intracardiac shunting may be mistakenly diagnosed. However, this pattern does not resemble any of the flow patterns described, and differentiation from true intracardiac shunting can usually be easily discerned. The overloading effect can be minimized by using lower gain and higher reject echographic settings.

Discussion Various agents including blood, saline solution and indocyanine green dye, when injected into the bloodstream, produce dense clouds of echoes. The explanations for this phenomenon include miniature bubbles in solution, turbulence, temperature differences and fluid acoustic impedance differences. 3,14 Any one explanation by itself does not appear to account for the dramatic contrast effect, and a summation of effects appears to be most likely. In our experience a more pronounced contrast effect was observed in the presence of low output states, increased hemoglobin and more forceful injection. We found that the manual injection of indocyanine green combined with a saline flush consistently produced a good contrast effect and was superior to any single agent. No toxic effect was obVolume 39

PERIPHERAL VENOUS CONTRAST ECHOCARDIOGRAPHY--SEWARD E-I" AL.

A

-Y"'-'-

I

FIGURE 10. Contrast echocardiograms. A, from a 2 year old boy with surgically proved common ventricle with outflow chamber and dextrotransposition of great arteries. The anterior leaflet of both the tricuspid (TV) and mitral (MV) valves is recorded simultaneously with no intervening septum. After peripheral venous injection of indocyanine green dye, the resultant cloud of echoes first appears in the tricuspid valve funnel (TVF) (arrowhead) and subsequently appears throughout the ventricular chamber. The mitral valve funnel (MVF) remains echo-free. B, from a 13 year old boy with single ventricle, two A-V valves and ievotransposed great arteries. With the transducer oriented to view the posteriorly !ocated A-V valve (mitral valve, [MV]) after hand venous injection of dye, the resultant echoes appear in diastole anterior to this A-V valve echo after the E point (arrows). The mitral valve funnel (MVF) remains echo-free.

served or has been reported as a result of intravenous or inadvertent subcutaneous injection of indocyanine green dye. 15-17 A few patients complained of mild local discomfort at the injection site coincident with the saline flush. Contrast echocardiography until recently was primarily applied to the verification of cardiac structures. Its use in the evaluation of blood flow patterns, shunts and valve incompetence has only recently been investigated. 2,4,5,s,ls-21 Valdes-Cruz et al. 22 did report the feasibility of utilizing peripheral veins for contrast echocardiography. However, all their injections were performed through large bore catheters positioned by means of a venous cutdown procedure. Our experience demonstrates that routinely established peripheral intravenous routes can be used to produce reproducible contrast echograms of good quality. R i g h t to left shunts: Contrast echocardiography appears to be a sensitive technique for detecting and localizing right to left shunting. Right to left shunts as small as 5 percent can be detected with contrast echographic techniques and may be reliably localized to the atrial, ventricular or pulmonary level. 4,5 It is reported

FIGURE 11. A 23 year old man with hypertrophic myopathy. After hand venous injection of dye, a dense cloud of echoes appears in the right ventricular outflow tract (RVO). Nearly simultaneously, echoes extend through posterior structures (arrows). These reverberations initially appear during diastole in the aortic root, inconsistent with any described flow pattern. AV = aortic valve; LA = left atrium.

February 1977

The American Journal of CARDIOLOGY

Volume 39

211

PERIPHERAL VENOUS CONTRAST ECHOCARDIOCIRAPHY--SEWARD ET AL.

that the majority of patients with an uncomplicated atrial septal defect will have a small right to left shunt. 23 In agreement with this observation, we were able to detect a small right to left shunt in patients with atrial septal defect whose overall shunt was predominantly left to right. Patients with an isolated ventricular septal defect, on the other hand, showed a right to left shunt only in the presence of increased right ventricular pressure (usually above two thirds of systemic pressure). In most patients with uncomplicated ventricular septal defect and predominant left to right shunt, we have not detected right to left shunt with echographic techniques. These observed echographic contrast flow patterns and their graphic demonstration of timing of right to left shunting are in agreement with the detailed angiographic work of Levin et al. 24-26 Right to left shunting at the atrial level (that is, atrial septal defect) occurs during ventricular systole with subsequent appearance of echoes in the left ventricle through the mitral valve orifice during ventricular diastole. Right to left shunting at the ventricular level (that is, ventricular septal defect) occurs predominantly in ventricular diastole corn-

mencing in the isovolumetric relaxation phase of the cardiac cycle.4,27 Tricuspid a t r e s i a and c o m m o n ventricle: Characteristic contrast echographic flow patterns in complex congenital cardiac defects (that is, tricuspid atresia and common ventricle) were appreciated with peripheral contrast echocardiography. In the appropriate clinical setting, a bedside standard M mode and contrast echocardiogram may be the most convincing confirmatory evidence for the diagnosis of tricuspid atresia or common ventricle before cardiac catheterization. Clinical application: Our study has established the feasibility and clinical usefulness of peripheral venous contrast echocardiography. We found that most superficial peripheral veins (hand, antecubital, foot) could be utilized. The resultant contrast echograms were reproducible and similar in quality to those obtained with more central (caval) injections. Bedside peripheral contrast echocardiography is a comparatively noninvasive technique for detecting and localizing right to left intracardiac or intrapulmonary shunting and an aid in the diagnostic assessment of certain congenital cardiac defects.

References 1. Gramlak R, Shah PM: Echocardiography of the aortic root. Invest Radiol 3:356-366, 1968 2. Gramlak R, Shah PM, Kramer DH: Ultrasound cardiography: contrast studies in anatomy and function. Radiology 92:939-948, 1969 3. Feigenbaum H, Stone JM, Lee DA, el al: Identification of ultrasound echoes from the left ventricle by use of intracardiac injections of indocyanine green. Circulation 41:615-621, 1970 4. Seward JB, TaJlk AJ, Spangler JG, et ah Echocardiographic contrast studies: initial experience. Mayo Clin Proc 50:163-192, 1975 5. Pleronl D, Varghese PJ, Rowe RD: Echocardiography to detect shunt and valvular incompetence in infants and children (abstr). Circulation 48:Suppl IV:IV-81, 1973 6. Feigenbaum H: Clinical applications of echocardiography. Prog Cardiovasc Dis 14:531-558, 1972 7. Feigenbaum H: Echocardiography. Philadelphia, Lea & Febiger, 1972, p 32 8. Shub C, Tajlk AJ, Seward JB, et ah Detecting intrapulmonary right-to-left shunt with contrast echocardiography: observations in a patient with diffuse pulmonary arteriovenous fistulas. Mayo Clin Proc 51:81-84, 1976 9. Meyer RA, Kaplan S: Echocardiography in the diagnosis of hypoplasia of the left or right ventricles in the neonate. Circulation 46:55-64, 1972 10. Van Praagh R, Ongley PA, Swan HJC: Anatomic types of single or common ventricle in man: morphologic and geometric aspects of 60 necropsy cases. Am J Cardiol 13:367-386, 1964 11. Lev M, Liberthson RR, Kirkpatrlck JR, et ah Single (primitive) ventricle. Circulation 39:577-591, 1969 12. Hallermann FJ, Davis GD, Rllter DG, et al: Roentgenographic features of common ventricle. Radiology 87:409-423, 1966 13. Seward JB, Tajik AJ, Hagler DJ, et ah Preoperative and postoperative echocardiographic observations in common ventricle (abstr). Circulation 52:Supp111:11-46, 1975 14. Kremkau FW, Gramlak R, Carstensen EL, et ah Ultrasonic detection of cavitation of catheter tips. Am J Roentgenol Radium Ther Nucl Med 110:177-183, 1970 15. Trlpp MR, Swayze CR, Fox IJ: Indocyanine green. In, Dye Curves:

212

February 1977

The American Journal of CARDIOLOGY

16. 17. 18. 19.

20. 21. 22. 23. 24. 25. 26. 27.

The Theory and Practice of Indicator Dilution (Bloomfield DA, ed). Baltimore, University Park Press, 1974, p 383-384 Cherrick GR, Stein SW, Leery CM, et ah Indocyanine green: observations on its physical properties, plasma decay, and hepatic extraction. J Clin Invest 39:592-600, 1960 Hunton DB, Bollman JL, Hoffman HN: Studies of hepatic function with indocyanine green. Gastroenterology 39:713-723, 1960 Kerber RE, Kioschos JM, Lauer RM: Use of an ultrasonic contrast method in the diagnosis of valvular regurgitation and intracardiac shunts. Am J Cardiol 34:722-727, 1974 Valdes-Cruz LM, Pleroni DR, Roland JM, et ah Detection of intracardiac right to left shunting by echocardiography (short communication). In, Ultrasound in Medicine (White D, Barnes R, ed). New York, Plenum Press, 1976, p 67-68 Valdes-Cruz LM, Pieroni DR, Roland J-MA, et ah Recognition of residual postoperative shunts by contrast echocardiographic techniques (abstr). Am J Cardiol 37:178, 1976 Duff DF, Gutgesell HP: The use of saline for ultrasonic detection of a right-to-left shunt in postoperative period (abstr). Am J Cardiol 37:132, 1976 Valdes-Cruz LM, Pieroni DR, Roland J-MA, et ah Echocardiographic detection of right to left shunts by peripheral vein injections (abstr). Circulation 52:Supp111:11-121, 1975 Rasmussen K, Simonsen S, Storsteln O: Quantitative aspects of right-to-left shunting in uncomplicated atrial septal defects. Br Heart J 35:894-897, 1973 Levin AR, Spach MS, Boineau JP, el al: Atrial pressure-flow dynamics in atrial septal defects (secundum type). Circulation 37: 476-488, 1968 Levin AR, Spach MS, Canent RV Jr, et al: Intracardiac pressureflow dynamics in isolated ventricular septal defects. Circulation 35:430-441, 1967 Levin AR, Boineau JP, Spach MS, et al: Ventricular pressure-flow dynamics in tetralogy of Fallot. Circulation 34:4-13, 1966 Assad-Morell JL, Seward JB, Tajlk AJ, el ah Echophonocardiographic and contrast studies in conditions associated with systemic arterial trunk overriding the ventricular septum: truncus arteriosus, tetralogy of Fallot, and pulmonary atresia with ventricular septal defect. Circulation 53:663-673, 1976

Volume 39