isolated two assessment and - heartity of anatomical substrates for congenital mitral regurgitation....

Br HeartJ 1982; 48: 109-16

Isolated anterior mitral cleftTwo dimensional echocardiographic assessment anddifferentiation from "clefts" associated withatnioventnicular septal defectJEFFREY F SMALLHORN, MARC DE LEVAL, JAROSLAV STARK, JANE SOMERVILLE,*JAMES F N TAYLOR, ROBERT H ANDERSON, FERGUS J MACARTNEY

Fromn the Thoracic Unit and Department ofPaediatric Cardiology, The Hospital for Sick Children, Great OmondStreet, and the National Heart Hospital, Westmoreland Street, London

SUMMARY Five patients with isolated clefts in the anterior leaflet of the mitral valve, unassociatedwith atrioventricular septal defects, are described. All had significant mitral regurgitation, with thecleft being the only abnormality in three. Two patients had an associated ventricular septal defect,one with a straddling right atrioventricular valve. Angiocardiography in four showed moderateregurgitation, but was not able to delineate the aetiology. Two dimensional echocardiographyshowed a constant defect in the anterior leaflet, pointing towards the left ventricular outflow tract.

This differed from 30 cases with atrioventricular septal defects where the "cleft" pointed towardsthe interventricular septum and was situated between the anterior and posterior bridging leaflets. Allcases with isolated clefts had surgical correction, with minimal residual regurgitation on follow-upexamination in two cases. Our current policy in patients with uncomplicated isolated cleft involvesnon-invasive assessment of these children and surgical correction if the regurgitation is significant.

Most cardiologists and surgeons have adopted a con-servative approach to the child with mitral regurgita-iion. This-is justified as the prospect of mitral valvereplacement in this age group is not to be consideredlightly, in view of the problems associated withprosthetic valves. I Reconstructive surgery may proveto be possible, but usually this cannot confidently bedecided before operation on the basis of traditionalinvestigational techniques. So the child goes for oper-ation in an atmosphere of uncertainty, with all thoseinvolved hoping, but usually not knowing, that recon-structive surgery will be possible.

Part of the basis for uncertainty is the wide variabil-ity of anatomical substrates for congenital mitralregurgitation. This may be produced by redundantand deficient chordae tendineae, parachute mitralvalve and its variants, anomalous mitral arcade,floppy mitral valve, and, rarely, Ebstein's anomaly ofthe mitral valve.2 3*Present address: National Heart Hospital, Westmoreland Street, London WIM8BA.Accepted for publication 29 April 1982

An isolated cleft in the anterior (aortic) leaflet of themitral valve is an uncommon cause of regurgitation,the cleft always pointing towards the left ventricularoutflow tract. But it is an important lesion because itis potentially correctable.4

In the past, M-mode echocardiography has pro-vided functional information about mitral regurgita-tion, but apart from mitral valve prolapse has yieldedlittle morphological detail as to the exact nature of thelesion.5 Angiocardiography has permitted recognitionof some specific varieties of congenital mitral valvedisease, but has proved singularly unsuccessful in therecognition of isolated mitral cleft.6 In contrast, twodimensional echocardiography should enable recogni-tion of an isolated cleft.We have treated a group of children where the

mitral regurgitation was the result of a localised cleftin the anterior leaflet, unassociated with atrioventricu-lar septal defects. We now describe their preoperativeassessment, surgical management, and our currentapproach to the problem.

109

on May 24, 2020 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.48.2.109 on 1 August 1982. D

ownloaded from

http://heart.bmj.com/

Smallhorn, de Leval, Stark, Somenrille, Taylor, Anderson, Macartney

Subjects and methods

All patients with isolated anterior mitral cleftsattended the congenital heart clinics of The Hospitalfor Sick Children, Great Ormond Street, or theNational Heart Hospital, Westmoreland Street. In allcases a clinical diagnosis of mitral regurgitation hadbeen made and radiological evidence of left ventricu-lar enlargement was present. Of the five patientsstudied, four had angiocardiographic evidence ofmoderate to severe mitral regurgitation. The fifthpatient, in view of our previous experience, was sub-mitted to operation without prior invasive studies.Two cases had an associated ventricular septal defect,one with a straddling right atrioventricular valvediagnosed by echocardiography. In the others, theanterior mitral cleft was an isolated defect.

All patients had both M-mode and two dimensionalechocardiographic assessments. The patients werestudied with either an Advanced TechnologyLaboratory mechanical sector scanner using a 3*0MHz transducer, or a Smith Kline mechanical scan-ner with a 2 25 MHz scan head. The apical and sub-costal four chamber views were used to assess theleaflets, as was the short axis view at the level of theaortic root. In the latter cut, a scan from the aorticroot to the apex of the left ventricle enabled an accu-rate assessment of the anterior and posterior mitralleaflets and associated papillary muscles.

In all patients M-mode studies were performed toassess the mitral valve motion and left ventricular cav-ity size. Two cases had preoperative, and three post-operative pulsed Doppler studies performed by sam-pling in the left atrium. All were submitted for surgi-cal repair of the cleft anterior mitral leaflet.To clarify the difference between an isolated

anterior mitral cleft and that associated with anatrioventricular septal defect, a further group of 30patients with the latter abnormality was included inthe study. All of these patients had either angiocar-diographic or surgical confirmation of the defect.

PREOPERATIVE ECHOCARDIOGRAPHIC STUDIES

Patients with an isolated mitral cleftIn all cases there was a localised cleft in the anterior(aortic) leaflet of the mitral valve (Fig. 1 and 2). Thiswas best appreciated in a short axis scan from theaortic root to the apex of the left ventricle. Just belowthe aortic root, a position is obtained wherein the leftventricular outflow tract and the area of aortic tomitral valvar fibrous continuity can be visualised.When the mitral valve is normal, the anterior leaflet isseen as a continuous dense echo (Fig. 3), but inpatients with an anterior cleft there is a break in thisecho (Fig. 4). In all cases the edges of the cleft

appeared as denser echoes suggesting that they wererolled and thickened, this observation beingconfirmed at operation (Fig. 4). During real-timestudy, an abnormal pattern of movement of theanterior leaflet existed throughout the assessment. So,during systole the two portions of the anterior leafletfloated together, whereas during diastole, the edges ofthe cleft parted widely. Most importantly, the cleftextended into the area of aortic-mitral continuity, andwas not directed towards the interventricular septum.All cases had two papillary muscles with no otherabnormality of the mitral valve being seen. In particu-lar, the chordae from the anterior mitral leaflet werenormally situated.

In the apical four chamber view the effective mitralorifice is frequently not visualised, as the anteriormitral leaflet crosses the plane of the transducer beambefore inserting into the posteromedial papillary mus-cle group. When a localised anterior cleft is present,an echo free area is readily visualised in this cut, rep-resenting the break in the leaflet. This appearance canalso be visualised in the subcostal four chamber view(Fig. 5).

In all cases the inferior limbus of the interatrialseptum was intact, and there was a normal atrioven-tricular muscular septum (Fig. 5). Thus, the atrioven-tricular valves were in all cases attached to the septum

Poste,o-med,al Cletl .- Anterior com.liswue 'Cleft' is gapA,eolaea conmm.ssuwe aort-C leaflet betweenAXXS5ea08,dg.g eales

Normcal Isolated cleft Atrioventriculakr defect

Fig. 1 A diagrammatic representation of the surgeon's view ofthe left atrioventnicular valve via a left atnial incision. Note thedifference between the cleft in a normal anteror mitral leaflet ascompared with the "cleft" seen in an atrioventricular septaldefect. In the latter case the "cleft" exists between the antenorand posterior bridging leaflets.

110



j.com/

Br H


ownloaded from


Two dimensional echo in mitral clefts

Fig. 2 Necropsy specimenfrom a patient with an isolatedcleft in the anterior mitralleaflet. The pictre on the leftshonvs the cleft anterior laflet.The picture on the nght showsthe cleft pointing towards thekft ventricular outflow tract.AML, anterior mitral leaflet;CL, cleft; LA, kft atrium;LV, kft ventricle; LVOT, kftventricular outflow tract. Thisfigure is reproduced by kindpernission ofProfessor AntonBecker, Amsterdam.

Fig. 3 The upperpanel shows a normal mitral valve at the kvel ofthe kft ventricular outflow tract. Note the anterior mitralleaflet is infibrous contidnuity with the aorta. The specimen on the right is cut in the same plane to indicate the echocardiographicfeatures. The lowerpanel is a lower short axis cut at the level ofthe mitral leaflets. Note thefishmouth appearance ofthe mitralvalve. The specimen on the right is cut on the same plane as the echocardiogram. MV, mitral valve; RV, right venticle;RVOT, right ventricular ou4ow tract; TV, tricuspid valve. See Fig. 2 for remaining abbrevtsatwns.

_,- ;LA

:;t AML

ill



j.com/

Br H


ownloaded from


Smallhorn, de Leval, Stark, SomerviUe, Taylor, Andlerson, Macartney

.

Fig. 4 The top picture is a short axiscut at the level of the left ventricularoutflow tract in a patient with anisolated cleft in the anterior mitralleaflet. Note the cleft, indicated by thetwo arrows pointing towards theoutflow tract, and not the rightventricle. The lower picture is a shortaxis cut at the level of the mitralvalve. Note again the cleft indicatedby the arrows, pointing towards theoutlet of the heart and not the rightventricle. IVS, intermentricularseptum. See Fig. 2 and 3 forremainting abbreviations.

at different levels, with the tricuspid valve beingattached more inferiorly. The membranous and inletsepta were intact in all patients, including the onewith a ventricular septal defect and straddiing right(tricuspid) atrioventricular valve. Surprisingly, thedefect in the latter patient was found to be in themuscular trabecular septum with chordae from theseptal tricuspid leaflet straddling to be attached in themorphologically left ventricle. In this case, previouslydescribed,7 a posterior interventricular septum goingto the crux was present. In the other case the defectwas perimembranous. M-mode assessments of themitral valve showed a normal "M" shaped patternwith some associated fluttering of the anterior leafletduring diastole. The remaining features were those ofatrial and ventricular enlargement.

The left ventricular end-diastolic volume wasgreater than two standard deviations above the meanfor age in every case and the left atrial size was alsosignificantly increased.

Patints with atrioventricular septal defectsIn all cases the short axis cut showed that the "cleft"was situated between the anterior and posterior bridg-ing leaflet pointing towards the interventricular sep-tum and not the left ventricular outflow tract (Fig. 6).

INTRAOPERATIVE FINDINGS AND MANAGEMENTOF ISOLATED MITRAL CLEFTSAll patients were operated on cardiopulmonarybypass, with moderate hypothermia and cold cardio-plegia. The left atrium was opened posterior to the

112



j.com/

Br H


ownloaded from



Fig. 5 The upperpanel is a subcostalfour chambercut in a case with an isolated cleft in the anteriormitral leaflet. Note the division in the anterior leafletindicated by the two arrows. The posterior mitralleaflet is seen in the upper part of the picture. Thelower picture is from the same patient shounng thepresence ofan atrioventricular muscular septumindicated by the two arrows. PML, posterior mitralleaflet; VAS, ventriculoatrial septum. See Fig. 2, 3,and 4 for remaining abbretiations.

Waterston groove. The mitral valve was tested byinjecting cold saline under pressure into the left ven-

tricular apical vent. The left atrium was moderatelyenlarged in all patients. A universal finding was a deepcleft in the anterior leaflet, pointing towards the aorticroot and extending to the annulus (Fig. 1 and 2). Theedges of the cleft were thickened and rolled. Theanterior part of the leaflet had chordal attachments tothe anterior papillary muscle. The subvalvularapparatus was normal in all patients.The repair consisted in all cases of closing the cleft

with interrupted sutures starting at the annulus a,ndprogressing towards the free edge of the valve. Theregurgitation at operation was completely abolished in

all cases but one, where a small central jet remained.In one patient in whom the regurgitation wasabolished at operation, on follow-up there was a softpansystolic murmur at the apex with no diastolicmurmur. In these two patients pulsed Dopplerultrasound showed a regurgitant jet in the left atrium.Three of the patients had no residual murmur ofmitral regurgitation, and this absence of regurgitationwas confirmed in one patient by pulsed Dopplerultrasound.

In the patient with the trabecular ventricular septaldefect and straddling right (tricuspid) atrioventricularvalve, the defect was closed through the tricuspidvalve, accepting a small defect around the papillary

113



j.com/

Br H


ownloaded from


Smallhom, de Leval, Stark, SomeniUe, Taylor, Anderson, Macarnw

Fig. 6 These pictures arefrom a case with an atrioventricular septal defect. The upper picture is a precordial short axis cut ata slightly higher level than that seen in the lowerpanel. Note that the "cleft" betteen the anterior and postenor bridging leafletspoints towards the interventricular septum and not the outflow tract of the left vntricle. The botom panel, at a lower level,again indicates the cleft between the leaflets pointing towards the interventricular septum. The specimen on the right isfrom apatient with an atrioventricular septal defect and patitoned orifices. The cut is in the same plane as the botom leftechocardiogram. Note the cleft, indicated by the arrow pointing towards the interventricular septum. LAV, left atrioventricularvalve; RAV, right atrioventncular valve. For remaining abbreviations see Fig. 3 and 4.

muscle attached to the left side of the septum.

Discussion

These findings throw light on two largely separatequestions, one to do with descriptive anatomy and oneclinical. Firstly, what is the difference between iso-lated cleft of the anterior mitral leaflet and the "mitralcleft" found in atrioventricular septal defects?

Secondly, how may patients with isolated cleft ofthe anterior mitral leaflet be distinguished frompatients with mitral regurgitations from other causes?

CLEFTS AND ATRIOVENTRICULAR SEPTALDEFECTSThe normal mitral valve consists of two leaflets, theanterior or aortic leaflet and the posterior or muralone. These,are divided by the anterolateral and pos-teromedial commissures which are supported bypaired left ventricular papillary muscles, theanterolateral and posteromedial group8 9 (Fig. 1)."Clefts" are usually described as part of the anatomyof atrioventricular septal defects. In these cases, anormal mitral valve does not exist and the left sidedvalve is formed from components of three of the five

114



j.com/

Br H


ownloaded from



basic leaflets, which may be found in all atrioventricu-lar septal defects. The "left sided cleft" in such heartsactually lies between the-anterior and posterior bridg-ing leaflets as they insert along the crest of the inter-ventricular septum'0 (Fig. 1 and 6).

Isolated clefts of the mitral valve are rare, but havepreviously been described in the anterior and pos-

terior leaflets unassociated with atrioventriculardefects.4 The cleft usually involves the anterior leafletand divides it in its entirety (Fig. 1). Anomalouschordal attachments are frequently a complicatingfeature though we found none in these patients. Itmust be stressed, however, that if the cleft involvesthe anterior leaflet and is to be regarded as isolated, itmust point towards theJleft ventricular outflow tract.In those with an atrioventricular septal defect,because of the very nature of the leaflet pattern the"cleft" can in no sense divide the anterior leaflet'0 asan isolated mitral cleft does. And because it does notdivide the anterior leaflet it points to the inter-ventricular septum (compare Fig. 1, 4, and 6) ratherthan to the left ventricular outflow tract. "IThese considerations lead us to suppose that, if the

very rare conditions of atrioventricular septal defectwith intact atrial and ventricular septa were encoun-tered, the cleft in the "left atrioventricular valve"would still point towards the septum. Furthermore,the other echocardiographic stigmata of atrioventricu-lar septal defects would still be present, notably thepresence of two atrioventricular valves at the same

level, resulting from absence of the atrioventricularseptum.'2 13

DIFFERENTIATION FROM OTHER MITRALANOMALIES, AND CONSEQUENCES FORMANAGEMENTMitral valve abnormalities in young children arealmost always of congenital origin., though mitralregurgitation may also be secondary to primary leftventricular disease, or to anomalous origin or atresiaof the left main coronary artery. In older children,rheumatic mitral disease is more common. Theechocardiographic features of congenital mitralstenosis have previously been described.'4 Inpatients in whom the predominant lesion is regurgita-tion, the majority have no gross malformation of thevalve, but possess redundant leaflets and a dilatedannulus. Others have grosser abnormalities such as

single papillary muscles. Over the past few years, par-ticularly since the advent of M-mode echocardiogra-phy, prolapse of the mitral leaflets has been suggestedas a likely cause of this regurgitation. Normally theposterior leaflet is involved, but occasionally theanterior one is. 16 17The majority of children with gross structural

abnormalities of the mitral valve who present in early

childhood have associated defects which frequentlymask the signs of mitral valve disease. Indeed, it israre for them to have only isolated mitral regurgita-tion. In the older age group cardiac failure is rare andthe lesion is usually detected during routine examina-tion. Even though mitral regurgitation is severe, thecombination of good left ventricular function and fre-quently massive left atrial enlargement ensure thatpatients maintain a relatively normal left atrial pres-sure and have little in the way of symptoms, at least intheir first decade. Because the condition is well toler-ated, and it has not been possible to guarantee thatsurgery will not involve mitral valve replacement, thetendency has been to postpone surgery until symp-toms demand it. Cardiac catheterisation andangiocardiography have in our experience often notprovided any more information than was obviousfrom clinical and non-invasive investigation.We believe that these results show the possibility of

a radical change in managment for at least one sub-group of patients with congenital mitral regurgitation.This is because two dimensional echocardiographyprovides the necessary detailed information as to theprecise anatomical abnormalities of the leaflets andsubvalvar apparatus. Such simple structural abnor-malities as localised clefts are readily visualised. Thesepatients can then be selected for early surgical repairon the basis that a good result can be anticipated,thereby avoiding the problems of left ventricular dys-function.which may supervene in patients with longstanding mitral regurgitation. Is We believe that thesepatients reflect the fact that the mitral regurgitation islargely, if not entirely, the result of the cleft. Repair ofthe cleft produces a normal or near normal anteriormitral leaflet (Fig. 7).

This situation is quite different from that pertain-ing to "left sided clefts" in atrioventricular septaldefects. Here the "cleft" is but one cause of leftatrioventricular valve regurgitation, and indeed maynot cause any regurgitation at all. Repairing the"cleft" in no way restores a normal "anterior mitralleaflet", and may actually result in left ventricularinflow obstruction.'0 19 20From a small series such as this one cannot be dog-

matic about the overall functional results of surgery.They do compare favourably, however, with theresults of mitral repair for all congenital mitralanomalies reported by Carpentier and colleagues.2'These authors found a persistent apical systolic mur-mur in 22 out of 34 patients, and a significant reduc-tion in heart size in only seven.Our present policy is therefore to perform two

dimensional echocardiographic studies in all patientswith clinical evidence of mitral regurgitation. If anisolated cleft is diagnosed and no other complicatingsuspected features are present, such as coarctation of

115



j.com/

Br H


ownloaded from


Smailhorn, de Leval, Stark, SomermUe, Taylor, Anderson, Macartney

Fig. 7 This is a postoperatiye short axis cutma patient with arepaired localised cleft in the anterior mitral leaflet. Note thesuture line indicated by the dense echo in the centre ofthe anteriormitral leaflet. See Fig. 2 and 3 for abbreviations.

the aorta or severe pulmonary vascular disease, thedegree of regurgitation is assessed clinically, radio-graphically, and by measuring left atrial and left ven-

tricular sizes. If it is graded as moderate to severe,then the patients are submitted for surgical repair ofthe valve. Careful follow-up studies of mitral valvefunction in a large series of patients will obviously benecessary.

JFS is a British Heart Foundation Junior ResearchFellow. RHA and FJM are supported by the BritishHeart Foundation and, respectively, the Joseph Levyand Vandervell Foundations.

References

1 Berry BE, Ritter DG, Wallace RB, McGoon DC, Daniel-son CK. Cardiac valve replacement in children. ThoracCardiovasc Surg 1974; 68: 705-10.

2 Davachi F, Moller JH, Edwards JE. Disease of themitral valve in infancy. An anatomic analysis of 55 cases.

Circulation 1971; 43: 565-79.3 Ruschhaupt DG, Bharati S, Lev M. Mitral valve mal-

formation of Ebstein type in absence of corrected trans-position. Am Cardiol 1976; 38: 109-12.

4 Goodman DJ, Hancock EW. Secundum atrial septaldefect associated with a cleft mitral valve. Br Heart1973; 35: 1315-20.

5 Schwartz DC, Kaplan S, Meyer RA. Mitral valve pro-lapse in children: clinical, echocardiographic and cine-angiographic findings in 81 cases (abstract). AmJ Cardiol1975; 35: 169.

6 Macartney FJ, Bain HH, Ionescu MI, Deverall PB, Scott0. Angiocardiographic/pathologic correlations in con-

genital mitral valve anomalies. Eur J Cardiol 1976; 4:191-211.

7 Smallhorn JF, Tommni G, Macartney FJ. Detectionand assessment of straddling and overriding atrioven-tricular valves by two-dimensional echocardiography. BrHeartJ 1981; 46: 254-62.

8 Lam JHC, Ranganathan N, Wigle ED, Silver MD.Morphology of the human mitral valve. I. Chordae ten-dineae: a new classification. Circulation 1970; 41: 449-58.

9 Ranganathan N, Lam JHC, Wigle ED, Silver MD.Morphology of the human mitral valve. II. The valveleaflets. Circulation 1970; 41: 459-67.

10 Carpentier A. Surgical anatomy and management of themitral component of atrioventricular canal defects. In:Anderson RH, Shinebourne EA, eds. Paediatric cardi-ology 1977. Edinburgh, London: Churchill Livingstone,1978: 477-90.

11 Beppu S, Nimura Y, Sakakibara H, et al. Mitral cleft inostium primum atrial septal defect assessed by cross-sectional echocardiography. Circulation 1980; 62: 1099-107.

12 Hagler DJ, Tajik AJ, Seward JB, Mair DD, Ritter DG.Real-time wide-angle sector echocardiography: atrioven-tricular canal defects. Circulation 1979; 59: 140-50.

13 Smallhorn JF, Tommasini G, Anderson RH, MacartneyFJ. Assessment of atrioventricular septal defects bytwo-dimensional echocardiography. BrHeartJ 1982; 47:109-21.

14 Driscoll DJ, Gutgesell HP, McNamara DG. Echocar-diographic features of congenital mitral stenosis. Am JCardiol 1978; 42: 25946.

15 Smallhorn JF, Tommasini G, Deanfield J, Douglas J,Gibson D, Macartney FJ. Congenital mitral stenosis.Anatomical and functional assessment by echocardiogra-phy. Br Heart J 1981; 45: 527-34.

16 DeMaria AN, King JF, Bogren HG, Lies JE, MasonDT. The variable spectrum of echocardiographic mani-festations of the mitral valve prolapse syndrome. Circula-tion 1974; 50: 33-41.

17 Bisset GS, Schwartz DC, Meyer RA, James FW, KaplanS. Clinical spectrum and long-term follow-up of isolatedmitral valve prolapse in 119 children. Circulation 1980;62: 423-9.

18 Kirklin JW. Replacement of the mitral valve for mitralincompetence. Surgery 1972; 72: 827-36.

19 Griffiths SP, Ellis K, Burris JO, Blumenthal S, BowmanFO Jr, Maim JR. Postoperative evaluation of mitral valvefunction in ostium primum defect with cleft mitral valve(partial form of atrioventricular canal). Circulation 1%9;40: 21-9.

20 Losay J, Rosenthal A, Castaneda AR, Bernhard WH,Nadas AS. Repair of atrial septal defect primum.Results, course, and prognosis. J Thorac Cardiovasc Surg1978; 75: 248-54.

21 Carpentier A, Branchini B, Cour JC, et al. Congenitalmalformations of the mitral valve in children. Pathologyand surgical treatment. J Thorac Cardiovasc Surg 1976;72: 854-66.

Requests for reprints to Professor F J Macartney,The Hospital for Sick Children, Great OrmondStreet, London WCIN 3JH.

116



j.com/

Br H


ownloaded from


isolated two assessment and - heartity of anatomical substrates for congenital mitral regurgitation....

Documents