DIAGNOSTIC METHODSECHOCARDIOGRAPHY-DOPPLER

Noninvasive evaluation of the ratio of pulmonary tosystemic flow in atrial septal defect by duplexDoppler echocardiographyAKIRA KITABATAKE, M. D., MICHITOSHI INOUE, M.D., MASATO ASAO, M.D.,HIROSHI ITO, M.D., TOHRU MASUYAMA, M.D., JUN TANOUCHI, M.D.,ToSHIO MORITA, M.D., MASATSUGU HORI, M.D., HIDEO YOSHIMA, M.D.,KENJI OHNISHI, M.D., AND HIROSHI ABE, M.D.

ABSTRACT The ratio of pulmonary to systemic flow (Qp/Qs) was noninvasively evaluated byduplex Doppler echocardiography in 22 patients with atrial septal defects (ASDs). Right and leftventricular stroke volumes (RSV, LSV) were determined from the recordings of ejection blood flowvelocity and diameter at the level of the pulmonary and aortic orifices in each ventricular outflow tract.The ratio RSV/LSV, determined by the duplex Doppler echocardiography, was compared with Qp/Qsby oximetry. The RSV/LSV for 10 normal subjects was 0.99 ± 0.05 (mean + SD), whereas the RSV/LSV for patients with ASD, 2.26 ± 0.63, was significantly higher than that for normal subjects (p <.01). In patients with ASD, a fairly good correlation was found between RSV/LSV and Qp/Qs (r.92, p < .01; y = 1.1 lx - 0.30), and this high correlation was found even in patients withcomplications such as pulmonary hypertension, mitral and tricuspid regurgitation, Eisenmenger com-plex, and ventricular septal defect. We also found that semilunar valve regurgitation modified the valueof RSV/LSV in accordance with the degree of regurgitation. These findings indicate that, with a fewlimitations, the Doppler index RSV/LSV is clinically useful in the estimation of the magnitude of theshunt flow in patients with ASD and that the limitations could be overcome by additional Dopplerexamination.Circulation 69, No. 1, 73-79, 1984.

CARDIAC OUTPUT can now be accurately measuredby quantitative Doppler echocardiography as a productof echocardiographic cross-sectional area of the as-cending aorta and Doppler aortic flow velocity integralover systole.'11 Recent articles4-6 have demonstratedthat cardiac output can be also determined by Dopplerquantitation of transmitral flow or pulmonary arterialflow with the use of echocardiographic or angiograph-ic views of cross-sectional area. Thus, we can nonin-vasively obtain the ratio of pulmonary to systemic flow(Qp/Qs) in various shunt diseases by first determiningpulmonary and systemic flow with quantitative Dopp-ler echocardiography and then dividing the former bythe latter.7 l' Aortic flow is usually satisfactorily deter-mined with little error; however, calculation of pulmo-

From the First Department of Internal Medicine, Osaka UniversityMedical School, The Department of Heart Center, Osaka PrefecturalHospital, Osaka, Japan.

Supported by Japan Heart Foundation Research grant for 1979, andgrant 57570349 from the Ministry of Education, Japan.

Address for correspondence: Akira Kitabatake, M.D., The First De-partment of Intemnal Medicine, Osaka University Medical School, 1-1-50 Fukushima, Fukushima-ku, Osaka 553, Japan.

Received Aug. 23, 1983; accepted Sept. 22, 1983.

Vol. 69, No. 1, January 1984

nary flow sometimes results in significant errors5 thatare mainly due to the inability to adequately visualizeand measure internal pulmonary arterial diameter fromthe echocardiographic image.5 13 In determining pul-monary arterial flow in patients with atrial septal defect(ASD) and/or ventricular septal defect, another seriousproblem arises in many cases; pulmonary arterial flowis often too highly disturbed to allow determination offlow velocity from Doppler recordings.'2.We studied a method for Doppler quantitation of

pulmonary flow in patients with ASD using Dopplerrecordings of the right ventricular outflow tract, inwhich flow is possibly less disturbed than in the pul-monary artery.12 We validated this method of deter-mining Qp/Qs in patients with ASD by comparingDoppler recordings of the right ventricular outflowtract with those of the pulmonary artery. Some addi-tional technical developments were also tested in themeasurement of valve orifice.

MethodsPatient selection. The study population consisted of 24 pa-

tients with ostium primum- or secundum-type ASD (11 male

73

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

KITABATAKE et al.

and 13 female patients ranging in age from 3 to 69 years. mean

30) who underwent diagnostic cardiac catheterization and 10normal subjects (eight men and two women ranging in age from37 to 59 years, mean 49) who served as control subjects. Two ofthe 24 patients with ASD were excluded because the diametersof their pulmonary orifices could not be determined by echocar-diography. In the other 22 patients all Doppler and echocardio-graphic recordings obtained were satisfactory. In the patientswith ASD Qp/Qs ranged from 0.51 to 4.6. Seven patients hadpulmonary hypertension (mean pulmonary arterial pressure

over 20 mm Hg). The presence or absence of associated valvu-lar heart diseases was determined by Doppler examination and/or contrast angiography. Patient characteristics and diagnosesare shown in table 1.

The interval between Doppler examination and cardiac cath-eterization was within 24 hr in 14 patients, within 1 week inthree patients, within 3 months in two patients, and within 6months in three patients. The clinical condition of the patientsdid not change between cardiac catheterization and Dopplerexamination.

Apparatus. We used a duplex Doppler echocardiograph (Hi-tachi EUB- 1 OB or Toshiba SDS-21A with SSH-41 A) for bothimaging and determination of flow velocity. Two-dimensionalechocardiographic images, M mode echocardiographic trac-ings, and Doppler-determined flow velocity patterns were ob-tained with the same transducer array (2.5 MHz for EUB-1OBand 2.4 MHz for SDS-21A). Any cursor line could be interro-gated for pulsed Doppler sampling and M mode echocardio-graphic tracing, and the ultrasound beam direction and the sam-

ple volume were monitored as a bright line and a spot on the linein the two-dimensional echocardiographic image. The sample

volume was a cylinder with a diameter of 3 mm and a length of 2mm at a depth of 7.5 cm for EUB-lOB and with a diameter of 5

mm and a length of 2 mm at a depth of 10 cm for SDS-2 1 A. Thepulse repetition rate was 5.4 kHz for the former and 6 kHz forthe latter. Doppler recordings were obtained in the Dopplermode, and sampling position was easily checked by switchingthe instrument from the Doppler to the real-time imaging mode.Doppler signals derived from structures were minimized by a

high-pass filter, and all signals were analyzed in real time by theChirp-Z transform for EUB-lOB and by the fast-Fourier trans-form for SDS-21A, and were simultaneously displayed on an

electrocardiogram, phonocardiogram, and M mode echocardio-gram at a paper speed of 100 mm/sec. The flow velocity direct-ed toward and away from the transducer was displayed aboveand below the baseline on the recording paper.

Ultrasound examinationMeasurements in the right venitricular outflow tract. Each

patient rested in a supine position and breathed in a relaxed wayduring Doppler examination. The transducer was placed, as

usual, in the third intercostal space at the left sternal border andwas angulated laterally to depict the right ventricular outflowtract and main pulmonary artery. Doppler recordings of flowvelocity were obtained with the Doppler beam directed as paral-lel as possible to the long axis of the right ventricular outflowtract and with the sample volume positioned just at the pulmo-nary orifice at the center axis of the right ventricular outflowtract (figure 1. A). We then advanced the sample volume to themain pulmonary artery and recorded the flow velocity pattern,which was then compared with that in the right ventricularoutflow tract. The Doppler incident angle and the diameter of

TABLE IPatients' characteristics and Doppler and catheter measurements

HR Doppler incident Doppler measurements Catheter

Patient Age (beats/ BSA angle (degrees) RSV LSV measurement ofNo. (yr) Sex Diagnosis min) (m2 RVOT LVOT (mli) (ml) RSVILSV Qp/Qs

1 43 F ASD, Eisenmenger 98 1.33 26 42 32.3 51.3 0.63 0.512 11 M ASD 72 1.45 0 23 89.7 75.5 1.19 1.153 5 M ASD, VSD 90 0.80 37 50 67.7 45.2 1.50 1.534 6 M ASD 73 0.64 0 26 61.2 38.0 1.61 1.575 39 M ASD, VSD, TR, MR, PR, PH 57 1.40 0 42 181.3 83.2 2.18 1.576 49 M ECD, TR, MR, PR, PH 57 1.67 0 40 158.0 75.1 2.10 1.757 66 F ASD, PR. PH 88 1.34 44 0 118.0 54.0 2.19 1.828 7 F ASD, PR 120 0.96 0 28 75.5 36.5 2.07 1.909 61 F ASD. TR, MSR, PH 63 1.43 0 23 82.1 46.0 2.33 2.0610 11 F ASD 60 1.38 25 32 110.8 52.7 2.10 2.0911 69 F ASD, MSR, PH 83 1.41 18 40 119.8 46.0 2.41 2.0912 22 F ASD 67 1.37 26 26 98.6 48.7 2.02 2.1013 7 F ASD 82 0.81 0 0 93.0 34.9 2.66 2.3014 47 F ASD 86 1.37 26 26 120.3 45.2 2.66 2.2015 19 F ASD 62 1.58 25 27 112.0 46.5 2.41 2.3116 15 M ASD 68 1.59 0 32 127.0 52.6 2.41 2.6717 35 M ASD 69 1.45 0 26 159.0 64.1 2.48 2.6718 31 F ASD 80 1.49 20 32 137.2 56.1 2.45 2.8819 3 M ECD, TR, PH 95 0.59 0 22 74.0 26.9 2.75 2.9020 43 M ASD 80 1.36 0 0 123.6 39.8 3.09 3.3621 59 M ASD 72 1.79 0 0 273.4 77.5 3.50 3.5522 9 F ASD 92 0.90 25 44 70.5 24.0 2.94 3.59

HR heart rate; BSA = body surface area; RVOT right ventricular outflow tract; LVOT left ventricular outflow tract; EisenmengerEisenmenger complex; VSD = ventricular septal defect; TR = tricuspid regurgitation; PR = pulmonary regurgitation; MR = mitral regurgitation; PH

pulmonary hypertension; ECD = endocardial cushion defect; MSR = mitral stenosis associated with regurgitation.

74 CIRCULATION

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

DIAGNOSTIC METHODS-ECHOCARDIOGRAPHY-DOPPLER

FIGURE 1. Measurements of the flow velocity anddialmeter in the right (A) and left (B) ventricular out-flow tracts of a 15-year-old boy with ASD (patient16). The Doppler beanm and the sample volume areshown as a white cursor line and a white dot on theline. respectively, in the midsystolic firozcn scani im-tae (A and B,eppeileft). fi-om which the Dopplerangle against the flow vectoi- is determiinled. [he di-ameter of the pulmlonary orifice is determined as aninterval between inner walls (solid white line in A,iuppie rig/it). The diameter of the aortic orifice isimieasured from the M onode echocardilogram (beamidiiection is displayed as a dashed linae in B, uippery/iht). A and B. botton, Doppler- recordings. RV

right ventricle; Ao = aorta. PA = pulmonary artery;LV left ventricle. SV = sample volume; PCGphonocardiogram; ECG = electrocardiogram.

the pulmonary orifice were measured from midsystolic echocar-diographic images. In the measurement of the diameter of thepulmonary orifice, we asked patients to rest in a left decubitusposition so that the pulmiionary orifice would appear as parallelto the ultrasonic beam as possible. For the real-time echocardilographic images midsystolic ininer diameter of the pulmonaryorifice was measured as an interval between the points at whichthe pulmonary cusps originated.

Measurement.s in the left vienitricular outflow tract. Aftermeasurements in the right ventricular outflow tract, the trans-ducer was moved onto the cardiac apex and angulated mediallyto depict the left ventricular outflow tract and the ascendingaorta in a left anterior oblique equivalent view (figure I. B). Wepositioned the sample volume just at the aortic orifice at thecenter axis of the left ventricular outflow tract and recorded theflow velocity pattern. We then advanced the sample volume intothe ascending aorta to record the aortic flow velocity for the

Vol. 69, No. 1, January 1984

evaluation of the presence of flow distui-bance. The Dopplerincidcent angle was determined from the midsystolic gated echocardiographic image. The midsystolic diameter of the aorticorifice was deterniined from M mode echocardiograms thatwere recorded with the ultrasound beam directed thiough theaortie orifice and perpendicular to the aortic root.Data analysis. On the assumption that the velocity profile

just at the semilunar valve orifice is flat, right and left ventricu-lar stroke volumes (RSV, LSV) were determined as products ofthe cross-sectional area and the velocity integral over systolewith the use of the following equations:

Stroke volume =7(D/2)9JfLVl/t) dt

Vl,;x(t) = C fdnia,(t) / (2focosfl)

where D is the midsystolic diametcr of semilunar valve orifice,ET is ejection time, V,,,j(t) is instantaneous maximum flow

75

-- hi ..... ...... ..... 1- . A-1

Al

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

KITABATAKE et al.

TABLE 2Flows disturbances detected hby DIoppler echocardiography

Nornial

RVO1PALVOTAorta

It) )10 - )1(1 1)1{ (-)

ASI)

22 ( -)

15 (-) 7 (+)22)(120 -). 2' (Nb)

Symbols ( + and - indlicate presencce or absence ot flow distur-hances.RVOT = right vcntricl[lali outflow tl-act, PA main pulmonaiay

atery; LVOT = left ventiKCulaIlr outflw tractl. NI' n otevIl\lutedL.

is instantaneous tm axitvelocity! eldil l(t) is'ltlNtlClS Ill muniiUl Doppli. .shif t fre-quency. fo is carrier frequency. c is soitnId velocity aindi)i0sDoppler incident angle against the long axis of ventrictular out-flow tract. Ihe integral of V,\,(t) was cletcti-ttied as an1 areaunder the envelope of the Doppler shift fi-equency pattern with adigitizer (Cat din SC). Contt-on). When the Doppler incident an-gle was less than 20 degrees there was no nieed to coirrect V0,(t)toI the angle. However, in the right and left ventiicular oLutflowtracts, respectively, the Doppler incident ancle failed to be lessthan 2() dcegrees in none (0%7c) and thice (30%/c) of 10 normalsubjects and nine (41' ) and 18 (X96/) 01 22 patients; in thesecases V1,,(t) was corrected for the angle. Both RSV and LSVwere calculated as averages of five consecutive cat diac cycles.Both ventriculat stroke indexes wete obtatned by dividinestIoke volumes by the body surface arca. The ratio RSV/LSVwas calculated for estiimating Qp/Qs.

In each iecord. the flow distui-bance was evaluated IriIi thcspectral width of systolic Dopplei frequcency pattei-n. I he flowdisturbance was considercdl to be piresent if the 12 dB spectralwidth was Icateri than 45 cii/scec.

Catheter measurements. All patiecnts undet-went cat ditaccatheterization. Hemodlynlamilic imecaSteI-CmeiClMtS WeIe made with aNo. 7F fluid-filled cathetei connectcd to a pressure tiansducer(Statham P23Db). OxygenI saturation in bloodl was meicasured inpulmiionary aiterial (Spa). pulmonary venous (Spv), systemicarterial (Ssa), anld mixed venlous (Smv) (the inferioi venia cavasample being weighed 3: 1 in combinationi with the superior venacava sample) samples. Ihe Qp/Qs was calculated fi-oTi theoxygen saturation in blood with the equation: Qp/Qs - (Ssa -

Smv)/(Spv Spa).Statistical analysis. Results are expriessed as mean. ± SD.

Student's t test was applied to coimipaie ventilculair stroke index-es for the patients with ASD aclnd nornmal subtccts. The Dopplerdeterminied RSV/LSV was compared with Qp/Qs by oximetrywith a simple linear regr-ession analysis.

Results

Interpretation of Doppler records. Ten normal subjectsand 22 patients with ASD were examined; the resultsare listed in table 2. None of normal subjects had aflow disturbance in the right ventricular outflow tractor main pulmonary artery. While none of the patientshad a flow disturbance in the right ventricular outflowtract, seven of 22 had pulmonary arterial flow distur-bances (figure 2). None of these patients or normalsubjects had a flow disturbance either in the left ven-tricular outflow tract or in the ascending aorta. On thebasis of these findings, each stroke volume was deter-mined with the flow velocity pattern and the diameterin the ventricular outtlow tract.

Right and left ventricular stroke indexes and the Dopp-ler-deternmined shunt ratio (RSV/LSV). The mean valuesof right and left ventricular stroke indexes for patientswith ASD were compared with those for normal sub-jects (figure 3). The left ventricular stroke index forpatients with ASD (40.6 ± 9.5 ml/m2) was not signifi-cantly different from that for normal subjects (46.1 +7.4 ml/mi). However, the right ventricular stroke in-dex for patients with ASD (88.8 -J- 26.9 ml/mn) wasdistinctly higher than that for normal subjects (45.8 ±7.2 ml/mi; p < .01).The Doppler-determined index RSV/LSV for the

normal subjects (0.99 * 0.05) indicated the equiv-alence of RSV and LSV. The RSV/LSV for patientswith ASD (2.26 ± 0.63) was significantly highercompared with that for normal subjects (p < .01, fig-ure 4).

Estimation of Qp/Qs with Doppler-determined RSV/LSV. In all 22 patients the Doppler-determined RSV/LSV was compared with Qp/Qs by oximetry. TheRSV/LSV correlated well with Qp/Qs (r = .92, p <.01; y = 1.llx - 0.30; figure 5). This good agree-ment was found even in patients with complicationssuch as pulmonary hypertension (Nos. 6, 9, 11, and19) and mitral and/or tricuspid regurgitation (Nos. 5,

TOWAM)

2KHZ.-

ECG

FIGLRE 2. Flow velocity patterns in the right ven-tricular outflow tract (left) and in the main putLmonaryaitery (r-ight) of a patient with high left to right shuntflow. The flow velocity pattern in the right ventricu-lar outflow tract represents a narrow spectrum insystole. which suggests absence of a flow distur-bance. On the other hand, the flow is too highlydisturbed in the main pulmonary artery to allow de-termination of the instantaneous maximal velocity.

CIRCULATION

TOWARD

2KHZ.

kl :. i#thai;

AWAY

ECG

76

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

DIAGNOSTIC METHODS-ECHOCARDIOGRAPHY-DOPPLER

RSImi/n2

NORMAL ASD NORMAL ASO

FIGURE 3. Doppler-determined indexes of right (left) and left (right)ventricular stroke index for patients with ASD and normal subjects. Theright ventricular stroke index (RSI) for patients with ASD is significant-ly higher (p < .01), whereas the left ventricular stroke index (LSI) iswithin the normal range.

6, 9, 11, and 19). In two patients with endocardialcushion defects (Nos. 6 and 19) the RSV/LSV wasapproximately the same as Qp/Qs. In a patient withASD associated with a ventricular septal defect (No. 3)the RSV/LSV of 1.5 showed a good agreement withthe Qp/Qs of 1.53. In a patient with Eisenmenger com-plex (No. 1) who had a dominant right-to-left inter-atrial shunt proved on cardiac catheterization, theRSV/LSV of 0.63 was distinctly lower than normal,but was comparable to the Qp/Qs of 0.51.

Semilunar valve regurgitation theoretically modifiesthe value of the Doppler-determined index RSV/LSV;thus, the effects of semilunar valve regurgitation on theRSV/LSV were extensively studied. In all normal sub-jects and patients with ASD, semilunar valve regurgi-tation was examined by Doppler echocardiographyand its degree was evaluated by observing the pointthat regurgitant signals reached on the two-dimension-al echocardiographic image. None of the patients hadaortic regurgitation, and only four with ASD had pul-monary regurgitation. In patients with mild pulmonaryregurgitation (regurgitation detected less than 1 cmfrom the pulmonary valve by Doppler echocardiog-raphy; patients 6 to 8), the RSV/LSV was similar to theQp/Qs. However, in a patient with moderate pulmo-nary regurgitation (regurgitation detected up to 2 cmfrom the pulmonary valve by Doppler echocardiog-

5

4.WI-

I--JInLU

XU

3.

2.

1

CJa

ASD group

* ASDO ASD,PR* ECD

0 ]ECD,PRA ASD,VSD

* A ASD,VSD,PR*ASD, Eisenmenger

M

*°MEAN ±S. 0.

_

Elk

NORMAL ASDFIGURE 4. The Doppler-determined RSV/LSV in normal subjects andpatients with ASD. The value of 2.20 + 0.80 for patients with ASD issignificantly higher than that for normal subjects (0.99 + 0.05). Abbre-viations as in table 1.

0 ASD /O ASD,PR

4 u ECD_ ECD, PR

X A ASD,VSD * /A ASD,VSD,PR0 *ASD,Eisenmenger

0

0 1 2 34

RSV/LSV by DOPPLERFIGURE 5. Qp/Qs determined by oximetry compared with Dopplermeasurement of RSV/LSV. There is a high correlation between Qp/Qsand RSV/LSV (r = .92, p < .01, n = 22). The regression equation(solid line) is y = 1. lx - 0.30. Symbols as in figure 4.

Vol. 69, No. 1, January 1984

.10 -

77

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

KITABATAKE et al.

raphy; patient 5) the RSV/LSV of 2.18 was higher thanthe Qp/Qs of 1.57.

Discussion

The usefulness of a pulsed Doppler echocardio-graphic technique in estimating Qp/Qs has been report-ed in young patients with a variety of congenital in-tracardiac shunt diseases such as ventricular septaldefect, common atrioventricular cannal, and tetralogyof Fallot.7 In these reports, the right and left ventric-ular output values were obtained from the measure-ments of pulmonary and aortic flow velocities anddiameters, respectively. However, patients with ASDfrequently have disturbed pulmonary flows due to highflow rates with or without dilated main pulmonaryarteries. Doppler recordings in this study (figure 2 andtable 2) revealed that seven of 22 patients had broad-ened spectral patterns on Doppler recordings of theirmain pulmonary arteries, which made it impossible todetermine the envelope of Doppler shift frequency. Incontrast, none of them had disturbed flow in the rightventricular outflow tract and this permitted us to obtainthe velocity integral throughout the ejection period.These results are comparable to the previous phonocar-diographic findings, 12 which demonstrate that flowdisturbances are more frequently exhibited in the mainpulmonary artery than in the right ventricular outflowtract in young-to-elderly patients with ASD. Thus, wepreferred to use the flow in the right ventricular out-flow tract rather than that in the main pulmonary arteryfor the determination of RSV.Flow velocity measured by the Doppler technique is

essentially the cosine function of the angle between theDoppler beam and flow vector, and thus the Dopplerincident angle should be taken into account when itexceeds a critical value because it might cause seriouserrors in determining flow velocity. In this study wefailed to set the Doppler incident angle at less than 20degrees (which causes 5% errors) in the right ventricu-lar outflow tracts of nine of 22 patients and in the leftventricular outflow tracts of 18 of 22 patients, althoughwe attempted to align the Doppler beam as parallel tothe long axis of each of the tracts as possible. Theseobservations differ from those of Sanders et al.7 Theyobserved that the Doppler beam was adequatelyaligned parallel to the flow when they measured pul-monary and aortic flow in children with a variety ofcongenital shunt diseases. The differences between ourresults and those of Sanders et al. might be explainedby the differences in the ages of the study populationsor possibly by the differences in the locations of shuntlesions. In adult patients with ASD, the right ventricle

often presents more marked dilatation than it does in

children because of the long persistance of the rightventricular volume overload. In such an instance, thealignment of the Doppler beam parallel to the flowvector would likely be difficult, especially in the leftventricular outflow tract. Thus, in patients with ASD,and especially in adult patients, the Doppler incidentangle should be taken into account in estimating notonly left but also right ventricular output.The major problem encountered in this investigation

was the difficulty in obtaining a clear echocardiograph-ic image of the pulmonary artery5 13; however, echo-cardiographically the pulmonary orifice could be more

clearly imaged. In this study, we referred to additionalreal-time echocardiographic images to locate the con-

junctive portion of pulmonary cusps to the orifice be-cause moving anatomic structures are more easily rec-

ognized on real-time than on stop-frame images.Furthermore, accurate measurement of pulmonary ori-fice diameter would be possible if the pulmonary ori-fice were axially depicted on the echocardiographicimages because vague images due to lateral resolutionwould be minimized. We kept the patients in a leftdecubitus position to reduce the covering effect of theleft lung and searched for another echocardiographicwindow to depict the pulmonary orifice more axially.Thus, inner diameter of the pulmonary orifice was

successfully measured axially in all but two patientswho were excluded from the study.

In the present study the RSV showed good agree-

ment with the paired measurements of the LSV innormal subjects. This observation confirms thefacts'4' '5 that bronchial arterial flow does not exceed2% of left ventricular output and that right ventricularoutput as determined by the dye-dilution technique iscoincident with Fick's calculation (difference less than15%). In patients with ASD, right ventricular strokeindex increased, but left ventricular stroke index was

not different from normal. These findings suggest thatthe systemic flow is maintained at normal levels inpatients with ASD. Although there is still controversyabout whether the systemic flow is decreased or not inadult ASD patients without heart failure,'6 1' it seemsthat the increase in pulmonary flow rather than thedecrease in systemic flow accounts for the increase ofQp/Qs in patients with ASD.

In the patient with Eisenmenger complex who had a

dominant right-to-left atrial shunt, RSV/LSV was lessthan 1 and was comparable to Qp/Qs. This findingindicates that our technique allows the evaluation ofthe right-to-left shunt as well as the left-to-right shunt,although further studies are needed to establish the

78 CIRCULATION

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

DIAGNOSTIC METHODS-ECHOCARDIOGRAPHY-DOPPLER

feasibility of this method for assessing Qp/Qs in pa-tients with right-to-left shunt.

The Doppler-determined index RSV/LSV showed agood agreement with Qp/Qs in patients with ASD;however, extracardiac shunt diseases and semilunarvalve stenosis and/or regurgitation might modify thevalue of RSV/LSV. The RSV/LSV in patients withpulmonary regurgitation is anticipated to overestimateQp/Qs, since pulmonary regurgitation produces theaugmentation of the Doppler measurement of RSV.The RSV/LSV of 2.18 was markedly higher than theQp/Qs of 1.57 in a patient with moderate pulmonaryregurgitation (regurgitant flow distributed up to 2 cmfrom the pulmonary valve), although we could esti-mate Qp/Qs from RSV/LSV with insignificant errorsin patients with minimum pulmonary regurgitation (re-gurgitant flow detected less than 1 cm from the pulmo-nary valve). In patients with aortic regurgitation, RSV/LSV might underestimate Qp/Qs for similar reasons.However, aortic regurgitation is rarely associated withASD, and none of our patients had aortic regurgitation.Also, these possible limitations could be overcome byperforming additional Doppler echocardiographic ex-aminations. RSV/LSV showed a good agreement withQp/Qs even in patients with mitral and/or tricuspidregurgitation, which indicates that the Doppler tech-nique permits us to estimate Qp/Qs in patients withmitral and/or tricuspid regurgitation.Our results demonstrate that the Doppler technique

allows the noninvasive evaluation of Qp/Qs with ahigh degree of accuracy and allows determination ofthe stage of ASD by the consecutive assessment ofshunt magnitude. Furthermore, this technique appearspromising in the evaluation of the surgical closure ofASD.

We gratefully acknowledge the excellent secretarial assis-tance of Miss Keiko Yoshikawa.

References1. Darsee JR, Walter PF, Nutter DO: Transcutaneous Doppler meth-

od of measuring cardiac output. II. Noninvasive measurement by

transcutaneous Doppler aortic blood velocity integration and M-mode echocardiography. Am J Cardiol 46: 613, 1980

2. Magnin PA, Stewart JA, Myers S, von Ramm 0, Kisslo JA:Combined Doppler and phased-assay echocardiographic estima-tion of cardiac output. Circulation 60: 388, 1981

3. Gisvold SE, Baubakk AO: Measurement of instantaneous bloodflow velocity in the human aorta using pulsed Doppler ultrasound.Cardiovasc Res 16: 26, 1982

4. Hoenecke HR, Goldberg SJ, Carnahan Y, Sahn DJ, Allem HD,Valdes-Cruz LM: Controlled quantitative assessment ofpulmonaryand aortic flow ratio by range gated Doppler in children withcardiac disease. Circulation 64 (suppl IV): IV-167, 1981

5. Goldberg SJ, Sahn DJ, Allen HD, Valdes-Cruz LM, Hoenecke H,Carnahan Y: Evaluation of pulmonary and systemic blood flow by2-dimensional Doppler echocardiography using fast Fourier trans-form spectral analysis. Am J Cardiol 50: 1394, 1982

6. Fischer DC, Sahn DJ, Friedman MJ, Larson D, Valdes-Cruz LM,Horowitz S, Goldberg SJ, Allen HD: The mitral valve orificemethod for noninvasive two-dimensional echo Doppler determina-tions of cardiac output. Circulation 67: 872, 1983

7. Sanders SP, Yeager S, Williams RG: Measurement of systemicand pulmonary blood flow and Qp/Qs ratio using Doppler and two-dimensional echocardiography. Am J Cardiol 51: 952, 1983

8. Valdes-Cruz LM, Mesel E, Horowitz S, Sahn DJ, Fisher DC,Larson D, Goldberg SJ, Allen HD: Validation of two-dimensionalecho Doppler for measuring pulmonary and systemic flows in atrialand ventricular septal defects. A canine study. Circulation 66(suppl II): 11-231, 1982

9. Sanders SP, Yeager S, Williams RG: Measurement of systemicand pulmonary blood flow and Qp/Qs ratio using echocardiogra-phy and Doppler velocimetry. Circulation 66 (suppl II): 11-231,1982

10. Meyer RA, Kalabathy A, Korfhagen JC, Kaplan S: Comparison ofleft to right shunt ratios determined by pulsed Doppler/2D-echo andFick method. Circulation 66(suppl II): 11-232, 1982

11. Stevenson JG, Kawabori I: Noninvasive determination of pulmon-ic to systemic flow ratio by pulsed Doppler echo. Circulation66(suppl II): 11-232, 1982

12. Lewis DH, Ertugrul A, Deitz GW, Wallace JD, Brown Jr JR,Moghadam AN: Intracardiac phonocardiography in the diagnosisof congenital heart disease. Pediatrics 23: 837, 1959

13. Griffith JM, Henry WL: An ultrasound system for combined cardi-ac imaging and Doppler blood flow measurement in man. Circula-tion 57: 925, 1978

14. Cudkowicz L, Abelmann WH, Levinson GE, Katznelson G, Jreis-saty RM: Bronchial arterial blood flow. Clin Sci 19: 1, 1960

15. Fritts HW Jr, Harris P, Chidsey CA III, Clauss RH, Coumand A:Validation of a method for measuring the output of the right ventri-cle in man by inscription of dye-dilution curves from the pulmo-nary artery. J Appl Physiol 2: 362, 1957

16. Popio KA, Gorlin R, Teicholtz LE, Cohn PF, Bechtel D, HermanNV: Abnormalities of left ventricular function and geometry inadults with an atrial septal defect. Am J Cardiol 36: 302, 1975

17. St. John Sutton MG, Tajik AJ, Mercier LA, Seward JB, GiulianiER, Ritman EL: Assessment of left ventricular function in secun-dum atrial septal defect by computer analysis of the M mode echo-cardiogram. Circulation 60: 1082, 1979

18. Goldberg SJ, Areias J, Feldman L, Sahn DJ, Allen HD: Lesionsthat cause aortic flow disturbance. Circulation 60: 1539, 1979

Vol. 69, No. 1, January 1984 79

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from

Yoshima and K OhnishiA Kitabatake, M Inoue, M Asao, H Ito, T Masuyama, J Tanouchi, T Morita, M Hori, H

by duplex Doppler echocardiography.Noninvasive evaluation of the ratio of pulmonary to systemic flow in atrial septal defect

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1984 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.69.1.73

1984;69:73-79Circulation. 

http://circ.ahajournals.org/content/69/1/73the World Wide Web at:

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

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

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

  document. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:

by guest on June 28, 2018http://circ.ahajournals.org/

Dow

nloaded from