DJ Radford, KR Bloom, T Izukawa, CA Moes and RD Rowepathological correlates

Echocardiographic assessment of bicuspid aortic valves. Angiographic and

ISSN: 1524-4539 Copyright © 1976 American Heart Association. All rights reserved. Print ISSN: 0009-7322. Online

72514Circulation is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX

1976, 53:80-85Circulation 

http://circ.ahajournals.org/content/53/1/80located on the World Wide Web at:

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

http://www.lww.com/reprintsReprints: Information about reprints can be found online at   journalpermissions@lww.comFax: 410-528-8550. E-mail: Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters  http://circ.ahajournals.org//subscriptions/Subscriptions: Information about subscribing to Circulation is online at

by guest on July 10, 2011http://circ.ahajournals.org/Downloaded from

Echocardiographic Assessment ofBicuspid Aortic Valves

Angiographic and Pathological CorrelatesDOROTHY J. RADFORD, M.B., M.R.C.P.(U.K.), KENNETH R. BLOOM, M.B., F.C.P.(S.A.).

T. IZUKAWA, M.D., F.R.C.P.(C). C. A. F. MOES, M.D., AND R. D. ROWE, M.D., F.R.C.P.(E)

SUMMARY Aortic root echocardiograms were recordedfrom 89 patients whose aortic valves had also been ade-quately defined by selective angiography or viewed sur-

gically or at autopsy. The eccentricity index (E.I.) of the aor-

tic leaflets was measured at the onset of diastole and an E.I.of 1.3 or greater was taken as abnormal.Of 31 patients with isolated nonobstructed or mildly

obstructed bicuspid aortic valves (7 viewed previously atvalvotomy and 24 diagnosed radiologically) 23 (74%) had an

abnormal E.I. Varying eccentricity occurred in some ofthese patients. Central leaflet echoes (E.I. of 1.0 to 1.25)were present in the other eight patients. All 14 patients withnonobstructed tricuspid aortic valves had central echoes.Additional multilayered diastolic echoes were found inpatients with bicuspid aortic valves as well as in two patients

BICUSPID AORTIC VALVE is the commonest congen-

ital abnormality of the heart. It occurs in 1.3 to 2.0% of rou-

tine autopsies.1 2

There is evidence that bicuspid valves may thicken, calcifyand result in aortic stenosis in adult life.3 These valves havea greater risk of being the site of bacterial endocarditis thando normal tricuspid aortic valves8 9 and are also prone todevelop incompetence."0 A risk of the associated sinuses ofValsalva becoming aneurysmal or rupturing has beenstated."1These possibilities of complications and the frequent in-

cidence of the condition in the population make it desirableto identify this abnormality by a reliable noninvasivemethod. Prophylaxis against infective endocarditis at timesof risk could then be advised and a follow-up program forearly detection of other complications could be instituted.

Clinically, an ejection systolic murmur and aortic ejectionclick may be heard. Whether these auscultatory signs are

specific for a bicuspid aortic valve and are present in every

case is not known.1'Echocardiography has developed rapidly as a reliable

noninvasive diagnostic tool. Reports of its use to distinguishbetween the normal and diseased aortic valves appear en-

couraging. 2, 13 A recent paper concluded that "echo-cardiography appears to be specific in the recognition of thebicuspid aortic valve.""

This present study was planned to assess the reliability ofechocardiography in the identification of isolated non-

obstructed or mildly obstructed bicuspid aortic valves.

From the Department of Pediatrics, Division of Cardiology, The Hospitalfor Sick Children, Toronto, Ontario, Canada.

Supported by a grant from the Ontario Heart Foundation.Address for reprints: Dr. Kenneth R. Bloom, Hospital for Sick Children,

555 University Avenue, Toronto, M5G IX8, Ontario, Canada.Received May 13, 1975; revision accepted for publication July 24, 1975.

with abnormal tricuspid aortic valves.The valves of 13 patients with aortic stenosis or in-

competence were viewed surgically and the E.I. was abnor-mal in all patients with a bicuspid aortic valve in this group.Aortic leaflet echo findings were not diagnostically helpful inten patients with tetralogy of Fallot, one of whom had a nor-mal E.I. with a surgically confirmed bicuspid aortic valve.Of 21 patients with VSD only one had a bicuspid aorticvalve but six had an abnormal E.I. This false positive signwas related to a high membranous VSD, sometimes withaortic valve prolapse.

It is concluded that an E.I. of > 1.3 in the absence of anassociated VSD is diagnostic of a bicuspid aortic valve andcan be expected to be found in approximately three-quartersof subjects with this abnormality.

Materials and Methods

Definition

A bicuspid aortic valve is one which has only twofunctional cusps. Anatomically there may be two discretecusps, two cusps with evidence of a small raphe on one, or

evidence of fusion of two cusps of an apparently tricuspidvalve such that there are only two functional segments.2'

Patients

A total of 89 patients whose ages ranged from 3 weeks to27 years comprises this series (table 1).

GROUP 1

Patients with a substantiated diagnosis of a nonobstructedor minimally obstructed bicuspid aortic valve withoutassociated intracardiac abnormalities were selected from thecomputerized records of the Cardiology Division of TheHospital for Sick Children, Toronto, and were recalled forechocardiographic assessment. All subjects had undergonecardiac catheterization including selective aortic rootanglograms.

A). Those patients with a firm radiological diagnosis of abicuspid aortic valve were accepted only if the measuredgradient of pressure across the aortic valve was less than 25mm Hg.

B). Patients with a surgically confirmed diagnosis of a

bicuspid aortic valve had undergone previous aorticvalvotomy. For inclusion, these patients had to have a

precise description of the aortic valve made at the time ofsurgery and a residual gradient across the valve of less than25 mm Hg on postoperative cardiac catheter study, or no

clinical, radiographic or electrocardiographic evidence of re-

stenosis.Groups 2, 3 and 4 of this series consist of patients who un-

80

by guest on July 10, 2011http://circ.ahajournals.org/Downloaded from

BICUSPID AV BY ECHO/Radford et al.

TAI3LE 1. Patient Groups

Numberof

Group patienits

1. Nonobstructed bicuispid aortic valvesA. Rt-adiologically diagiiosed 24

(i) with otherwise inorinal cat-heter study 10(ii) with coaretatioii 6(iii) with mild stenosis or ineompetenice 8

B. Surgieally diagnosed at auortic valvotomy 72. Nonobstructed tricuspid aortic valves

1Rtadiologically diagnosed 14(i) with nournal catheter study 6(ii) with coarctationl l(iIi) with mild steniosis 2(iv) with enidocardial fibroelastosis 1

3. Aortic valve disease viewed at surgery 13(i) bicuspid aoi tic valve 7(ii) tricuspid aortic valve 6

4. Other intracardiac lesioIIs 31A. Venitricular septal defects 21B. Tetralogy of Fallot 10

Total patients assessed 89

derwent cardiac catheterization at this hospital during theperiod July 1974 to January 1975. All were studied by echo-cardiography but only those who had adequate aortic rootangiograms were reviewed.

GROUP 2

Patients with nonobstructed tricuspid aortic valves. Thesesubjects had no associated intracardiac malformations, agradient of less than 25 mm Hg and a radiologicallytricuspid aortic valve.

GROUP 3

Patients with congenital aortic valve stenosis, or subvalvestenosis or aortic regurgitation. All in this group had subse-quent confirmation of the aortic valve appearances at sur-gery, and thus enabled us to make further visual-echo-graphic correlates.

GROUP 4

Patients with other cardiac anomalies but adequatelyassessed aortic valves (by radiology and/or surgery). Thisgroup consists of children with (A) ventricular septal defectsand (B) tetralogy of Fallot.

Radiology

Only patients who had good quality selective aortic rootangiograms, including a lateral or left anterior oblique view,were selected. It was essential for the catheter to be placed inthe aortic root close enough to the valve to allow gooddelineation by injected contrast material, but not so close asto distort the valve appearances. The radiological diagnosisof a bicuspid aortic valve was made on the diastolicappearances of the aortic sinuses and the systolicappearances of the valve leaflets similar to the description bySimon and Reis"5 for stenosed bicuspid valves. Wedocumented diastolic valve appearances as follows: (a) threeequal sinuses, (b) three unequal sinuses, (c) two unequalsinuses, and (d) two apparently equal sinuses. In addition,

HIGURE I Angiograms of a normal tricuspid aortic valve in thelateral projection (left) in diastole showing three sinuses of ap-parently slightly unequal size. (The noncoronary sinus is the lowestin this projection.) Right) In systole showing three cusps opening.

the systolic appearances of the leaflets, and evidence of com-plete cusp fusion, were recorded. Normal tricuspid aorticvalves had sinuses classified as (a) or (b) and three leaflets.Bicuspid aortic valves had sinuses classified as (b), (c) or (d)and the systolic appearances of movement of only two cusps,one of which might be larger than the other. Since there isoverlap of the diastolic appearances of the classification (b)to both groups, the systolic appearances are important (figs.1, 2 and 3).

Echocardiography

A commercially available Ekoline 20 ultrasonoscopeinterfaced to a Cambridge strip chart recorder was used witha variety of transducers varying from 2.25 MHz ½/2 inchdiameter focused at 5 centimeters to 5 MHz 14 inch non-focused. Echocardiograms of the aortic valve were recordedby first identifying the anterior leaflet of the mitral valve andthen rotating the transducer superiorly and medially untilthe aortic root was identified. Transducer position was thenvaried until the opening and particularly the closingmovements of the aortic leaflets were recorded. The instru-ment sensitivity was adjusted to produce clear leaflet echoes.This maneuver was repeated several times using different

FIGURE 2 Angiograms of a bicuspid aortic valve (left) in diastoteshowing three unequal sinuses and (right) in systole showing onlytwo cusps. The right and left cusps are fused. The aortic root isdilated.

81

by guest on July 10, 2011http://circ.ahajournals.org/Downloaded from

Vol. 53, No. 1, January 1976

|1~~~~~I II1 IIIIIIII111 IIIItlllllll'I I 1Onset of DiostoleIt t I I i I

FIGURE 3 A bicuspid aortic valve angiogram (left) in diastolewhen two sinuses are seen and (right) in systole when two cusps are

seen.

transducer angulations. Measurements of the "eccentricityindex" (E.I.) of the closed aortic valve leaflets were madespecifically at the onset of diastole (see discussion). The E.I.was calculated as follows:"

E..

=t/2 width of aortic lumenE.I minimum distance of the cusp echo to nearest

aortic marginThe internal diameter of the aortic root was used and bothmeasurements were taken in millimeters at the onset ofdiastole (fig. 4). In each patient at least three measurementswere taken. Where wide variation was noted the maximumE.I. was recorded and used in the assessment. Those with"central echoes" had an E.I. of 1.0 to 1.25 maximum. AnE.I. of 1.3 or greater was taken as abnormal (fig. 5). Thiscontrasts with a previously published series" in which an un-

defined zone occurred between tricuspid aortic valves withan E.I. of 1.0 to 1.25 and bicuspid aortic valves with an E.I.of 1.5 to 5.6. The presence of additional multilayered echoeswithin the aortic root was also recorded. All the echo-cardiograms were reported routinely, but reviewed again byone person for consistency. Only those in which the diastolicclosure of the cusps and anterior and posterior aortic wallswere clearly recorded were used.

Surgery

We attempted to obtain as much correlation of echofindings with visualized anatomy as possible. Groups 3 and 4were therefore included in the study for this purpose. Thedescription of the aortic valve was obtained from the sur-

geon or from the surgical records, which usually included a

drawing of the aortic valve if it was abnormal.

Results

Groups 1 and 2

The aortic root echo findings for the patients withnonobstructed aortic valves are shown in table 2. Eccentricaortic valve leaflets were recorded in 18 of 24 patients ingroup IA considered to have a bicuspid aortic valve on

radiological assessment (75%) and in five of seven patients in

FIGURE 4 A normal aortic root echocardiogram showingmeasurement of the eccentricity index at the onset of diastole:Eccentricity Index

Aortic root diameter at onset of diastole X l'2 A X ½2

Smallest distance a

group 1B proven to have a bicuspid aortic valve at surgery

(71%) thus giving an abnormal E.I. in 74% of patients with a

nonobstructed bicuspid aortic valve.In group IA the range of values for abnormal E.I. in the

18 patients was from 1.3 to 2.7. The average value was 1.7.The six patients with "central echoes" had an average valueof 1.1 and range 1.0 to 1.2. In group I B, five patients had an

5d <,W~~~- 5~-r - ;.-vr~. - v~. *7- ,,$. .--..f.-...

FIGURE 5 The aortic root echocardiogram of a patient with abicuspid aortic valve. The eccentricity index is 1.3 posteriorly.

82 CIRCULATION

by guest on July 10, 2011http://circ.ahajournals.org/Downloaded from

BODY AND ELECTRODE POSITION AND VCG/Riekkinen, Rautaharju

TABLE 2. Aortic Root Echo Findings in Patients withNonobstructed or Minimally Obstructed Aortic Valves

Number of patientsEcho features Group 1A Group 1B Group 2

Eccentricity index _ 1.3 18 5 0(Additional multilayered echoes) (3) (3) (0)

Central leaflets(Index 1.0 - 1.25) 6 2 14(Additional multilayered echoes) (1) (0) (2)

Total patients in group 24 7 14

E.I. ranging from 1.4 to 2.3 with an average of 1.8. Themaximum values for E.I. recorded in the other two with sur-

gically proven bicuspid valves were 1.1 and 1.2, respectively.Figure 6 shows the echoc"ardiogram of one of these latterpatients with central leaflet echoes. No significant eccen-

tricity was found on any view of the aortic root with alteredtransducer positions. Also, the echo recordings of the eightpatients with normal E.I. (six from group IA and two fromI B) were reviewed to see if there was a change in leaflet posi-tion at any point in diastole. None showed changing of posi-tion or more marked eccentricity.

All 14 patients with a tricuspid aortic valve on radio-logical assessment (group 2) had a normal E.I. (1.0 to 1.25).One patient in this group had endocardial fibroelastosis anddied. Postmortem confirmation of a tricuspid aortic valvewas obtained.

Additional multilayered echoes were found in each groupincluding two patients with tricuspid aortic valve. One ofthese patients had mild aortic stenosis with a measuredgradient at catheterization of 21 mm Hg. The other patientdied following rupture of an infected anastomotic site aftercoarctectomy. His aortic valve was shown at autopsy to betricuspid but with thickened and ridged leaflets (fig. 7).

Group 3

The surgically confirmed diagnoses together with theechocardiographic and clinical findings of the patients withsignificant aortic valve disease are in table 3. It will be notedthat all patients with a bicuspid aortic valve had an eccen-

tricity index of greater than 1.3 and all with a tricuspid aor-

tic valve had an index of less than 1.3.

Group 4A

Of 21 patients with ventricular septal defects (VSD) whohad satisfactory angiography of the aortic root, six (29%)had eccentric aortic leaflets (E.I. 2 1.3) by echocar-diography. Of these patients, one who had a muscular VSDwas considered to have a bicuspid aortic valve radio-

0ll l l I I I

I* X

x

*.ve..eq.Y-.M

*.

mIArnfm1uffirnlMllllmifirWlMllmrffrrxfiBiMiumTrnfr

FIGURE 6 The aortic root echocardiogram of a patient whoseaortic valve was viewed at surgery and observed to be bicuspid withright and left cusps. The echo of the aortic leaflets is central.

logically. The other five were similar in that they had highmembranous defects and tricuspid aortic valves on angiog-raphy. Four of these patients had radiological evidence ofaortic cusp prolapse through the VSD (two with aortic in-competence). The fifth patient had no definite evidence ofprolapse.The remaining 15 patients in the total group of 21 with

VSD had central aortic leaflets by echocardiography andconfirmation of tricuspid aortic valves on angiography orsurgery. These included two patients with aortic regurgita-tion and prolapse of the right coronary cusp through a highmembranous VSD.

Group 4B

None of the ten patients with tetralogy of Fallot had anabnormal eccentricity index (_ 1.3). However, one was con-sidered to have a bicuspid aortic valve by angiography andthis was confirmed at surgery. This patient's E.I. was 1.2.Another patient had radiological evidence of prolapse of the

TABLE 3. Findings in Patients in Group 3 (With Aortic Valve Disease Viewed Surgically)

Number with Nu1mber withNumber of eccentricity Number with multilayered

Diagnosis patients index _ 1.3 central index echoes

Valvular aortic stenosis with bicuspid aortic valve 6 6 0 3Gross aortic incompetence with bicuspid valve 1 1 0 0Valvular aortic stenosis with tricuspid aortic valve 1 0 1 0Aortic incompetence with tricuspid aortic valve 1 0 1 0SuLbvalvular and valvular.aortic stenosis with 4 0 4 0

tricuspid aortic valve

83

by guest on July 10, 2011http://circ.ahajournals.org/Downloaded from

Vol. 53, No. 1, January 1976

Abnormal Tricuspid Aortic Valve

II!f 1 , l

FIGURE 7 Left) Echocardiogram of a boy with coarctation of the aorta. The closed aortic leaflets are central but thereare additional multilayered diastolic echoes. Right) Autopsy photograph of the same patient. The aortic valve is tricuspidbut the leaflets are thickened and ridged.

right aortic cusp and mild aortic regurgitation. His E.I. wasalso 1.2. One patient with a central echo and a tricuspid aor-tic valve by angiography had autopsy confirmation of a nor-mal tricuspid aortic valve.

Discussion

Normal aortic valves are tricuspid and have equal-sizedcusps with commissures situated equally at one third of thecircumference of the aortic ring, resulting in three sinuses ofequal size. Variations occur in which there are three sinusesof unequal size and three unequal cusps without fusion.6Bicuspid aortic valves may have three unequal sinuses andevidence of fusion of two cusps with a wide raphe visible, orthere may be only two sinuses, two commissures and a smallraphe.2 Other variations of the anatomy of aortic valves arerarer and include unicommissural unicuspid valves, acom-missural valves, and quadricuspid valves.1', 17 It is importantto be aware of these anatomical variations when endeavor-ing to study the aortic valve by angiography or echocar-diography.The angiographic diagnosis of a bicuspid aortic valve can-

not be considered totally reliable, because there arevariations in appearances.18 For this reason we haveattempted to obtain visual confirmation of the echo andangiographic appearances whenever possible to minimizethe possibility of error when one image technique is com-pared with another. For greater reliability a small group ofseven patients whose aortic valves were seen at surgery andnoted to be bicuspid, and who currently have no evidence ofre-stenosis, have been included. In addition, the valves of allpatients with significant aortic valve lesions have beenvisualized at surgery, and a small number of patients in eachof the other groups have had follow-up of the radiologicaland echocardiographic assessments at surgery or autopsy.Abnormal eccentricity of the aortic leaflets on echo-

cardiography has been found in 74% of our patients whohave nonobstructed bicuspid aortic valves, and in no patient

with a tricuspid aortic valve, except where there is anassociated high membranous VSD usually with aortic cuspprolapse. It would therefore appear that an eccentricity in-dex of _ 1.3 is diagnostic of a bicuspid aortic valve if there isno associated VSD. However, the reverse is not true in thatcentral aortic leaflet echoes have occurred in both bicuspidand tricuspid aortic valves. This fact has been confirmed byvisualization of the valves at surgery.

Multilayered echoes in the aortic root, predominantly indiastole, have been observed in our study and described byothers previously.14 We observed this sign with the instru-ment sensitivity set to give clear aortic cusp echoes andfound that only additional nonspecific ultrasonic "noise"was produced by further increase in gains. This sign ofdiastolic multilayering has not been specific for bicuspid aor-tic valves in our series, but has indicated that the aortic valveis abnormal, as confirmed at autopsy in one case.The main exception to the rule that an eccentricity index

of _ 1.3 indicates a bicuspid aortic valve has been when thereis an associated high membranous VSD. However, this falsepositive sign of bicuspid aortic valve is not present in everycase of aortic valve prolapse through a VSD, so that it can-not in any sense be regarded as diagnostic of that clinicalsituation.

For assessment of E.I., we have obtained adequate re-cordings of the closing movements of the aortic leaflets andthen made the measurements specifically at the onset ofdiastole at a precise and constant point. This varies from themethod of Nanda et al.,4 which we found to give falsepositive diagnoses in a few instances. When the closingmovements of the leaflets are shown to merge into a line indiastole, it ensures that a true echo of the closed leaflets isrecorded. Otherwise, eccentric linear echoes that reallyoriginate from the aortic ring, wall or sinuses may be falselyinterpreted to be coming from the aortic valve leaflets. Wemade our recordings from several transducer positions andwhen variations occurred, documented the maximum E.I. Inno case were the leaflets central at the onset of diastole and

84 CIRCULATION

by guest on July 10, 2011http://circ.ahajournals.org/Downloaded from

BODY AND ELECTRODE POSITION AND VCG/Riekkinen, Rautaharju

eccentric in late or end-diastole. Since the undefined valuesof E.I. of 1.25 to 1.5 in a previous paper14 are an importantzone, we made a cut-off value of 1.3 for abnormal eccen-tricity. From the observed values, a lower figure would notinclude all with a bicuspid valve, so some false negative casesappear inevitable. It has been possible to obtain suitablerecordings in almost all patients.We have encountered beat to beat variation in eccentricity

at the onset of diastole, recorded from a constant transducerposition. These variations occurred at a more rapid rate thanrespiration. Thus, they may be related to the forwardsystolic movement of the aortic root during the cardiac cycleresulting in the transducer beam transecting the root atdifferent angles during this motion. Nanda et al.'4 observedvariations of leaflet position throughout one diastole andrelated this finding, and also the appearance of multiplediastolic echoes, to redundant folds of leaflet tissue.

The study of Glancy et al.'9 indicated that abnormal aor-tic valves occur more frequently with tetralogy of Fallotthan in the general population. We have not found echo-cardiography to be useful diagnostically in our small groupof patients with tetralogy of Fallot whose aortic cusps wereadequately recorded. E.I.s of 1.2 occurred in tetralogypatients with normal tricuspid aortic valves, and also in aproven bicuspid aortic valve and in a known case of aorticvalve cusp prolapse through the VSD. The difficulties withechocardiographic diagnosis in this group may be related tothe fact that the aortic root is large and anteriorly situatedand there is an associated high VSD.

In conclusion, echocardiography of the aortic root isuseful in making the diagnosis of a bicuspid aortic valvewhen the E.I. is . 1.3, provided that there is no associatedhigh membranous VSD. However, a normal E.I. does notexclude the presence of a bicuspid valve since central aorticleaflet echoes occur in approximately 25% of cases. Normaltricuspid aortic valves without associated VSD all have acentral E.I. Multilayered echoes in diastole occur with bothbicuspid and tricuspid valves, but this sign indicates that thevalve is abnormal. Echocardiographic analysis of the aortic

root is an important screening procedure in the detection ofbicuspid aortic valves.

Acknowledgment

We thank Miss Connie Williams for technical assistance and Mrs. JennyVisanji for secretarial work.

ReferencesI. Koletsky S: Congenital bicuspid valves. Arch Intern Med 67: 129, 19412. Roberts WC: The congenitally bicuspid aortic valve. A study of 85 autop-

sy cases. Am J Cardiol 26: 72, 19703. Smith DE, Matthews MB: Aortic valvular stenosis with coarctation of

the aorta with special reference to the development of aortic stenosisupon congenital bicuspid valves. Br Heart J 17: 198, 1955

4. Bacon APC, Matthews MB: Congenital bicuspid aortic valves and theetiology of isolated aortic valvular stenosis. Q J Med 28: 545, 1959

5. Campbell M: Calcific aortic stenosis and congenital bicuspid aorticvalves. Br Heart J 30: 606, 1968

6. Roberts WC: The structure of the aortic valve in clinically isolated aorticstenosis: An autopsy study of 162 patients over 15 years of age. Circula-tion 42: 91, 1970

7. Schlant RC: Calcific aortic stenosis. Am J Cardiol 27: 581, 19718. Osler W: The bicuspid condition of the aortic valves. Trans Assoc Am

Physicians 1: 185, 18869. Lewis T, Grant RT: Observations relating to subacute infective endocar-

ditis. Heart 10: 21, 192310. Edwards JE: The congenital bicuspid aortic valve. Circulation 23: 485,

196111. Wood P: Diseases of the Heart and Circulation, ed 3. London, Eyre and

Spottiswood, 1968, p 39012. Gramiak R, Shah PM: Echocardiography of the normal and diseased

aortic valve. Radiology 96: 1, 197013. Feizi 0, Symons C, Yacoub M: Echocardiography of the aortic valve. Br

Heart J 36: 341, 197414. Nanda NC, Gramiak R, Manning J, Mahoney EB, Lipchik EC, Deweese

JA: Echocardiographic recognition of the congenital bicuspid aorticvalve. Circulation 49: 870, 1974

15. Simon AL, Reis RL: The angiographic features of bicuspid and unicom-misural aortic stenosis. Am J Cardiol 28: 353, 1971

16. Falcone MW, Roberts WC, Morrow AG, Perloff JK: Congenital aorticstenosis resulting from a unicommissural valve. Clinical and anatomicfeatures in 21 adult patients. Circulation 44: 272, 1971

17. Hurwitz LE, Roberts WC: Quadricuspid semilunar valve. Am J Cardiol31: 623, 1973

18. Roberts WC, Elliott LP: Lesions complicating the congenitally bicuspidaortic valve. Anatomic and radiographic features. Radiol Clin North Am6: 409, 1968

19. Glancy DL, Morrow AC, Roberts WC: Malformations of the aorticvalve in patients with tetralogy of Fallot. Am Heart J 76: 755, 1968

85

by guest on July 10, 2011http://circ.ahajournals.org/Downloaded from