S endocardial cushion defects, with the incidence ranging. Surgical Options in Subaortic Stenosis Associated With Endocardia1 Cushion Defects

Transcription

1 Surgical Options in Subaortic Stenosis Associated With Endocardia1 Cushion Defects Serafin Y. DeLeon, MD, Michel N. Ilbawi, MD, William R. Wilson, Jr, MD, Rene A. Arcilla, MD, Otto G. Thilenius, MD, Saroja Bharati, MD, Maurice Lev, MD, and Farouk S. Idriss, MD The Heart Institute for Children, Christ Hospital and Medical Center, Oak Lawn, Illinois, and Department of Surgery, Children s Memorial Hospital and Northwestern University Medical School, Chicago, Illinois Over a 15-year period, 1 patients with endocardial cushion defects undergoing correction had subaortic stenosis requiring operative intervention. Ages ranged from 4 months to 17 years (mean, 7 6 years) and subaortic gradients from 15 to 100 mm Hg (mean, 60 f 5 mm Hg). Subaortic stenosis was due to discrete fibromuscular tissues in 7 patients, mitral valve malattachment in 3, and tunnel outflow in. In, the subaortic stenosis was clinically significant at the time of endocardial cushion defects repair, whereas in 10 it was noted to 14 years postoperatively (mean, 6.3 f 5 years). Surgical relief of subaortic stenosis was accomplished by resection of muscle tissues in 7, apicoaortic conduit insertion in, modified Konno procedure (aortic valve preserved) in, and lifting of malattached mitral valve from the outflow in 1. There was no early death and one late death (infected conduit). Severe mitral insufficiency developed in the patient who had the mitral valve lifted and necessitated valve replacement. Postoperative echocardiographic gradient in 9 patients ranged from 0 to 36 mm Hg (mean, mm Hg). Clinically significant subaortic stenosis has not developed in any patient in 15 years of follow-up (mean, 5 4 years). We conclude that in subaortic stenosis associated with endocardial cushion defects, resection is effective for discrete obstruction, whereas a modified Konno procedure is preferable for obstruction due to tunnel outflow or mitral valve malattachment. (Ann Tlzorac Surg 1991;5: ) ubaortic stenosis is not uncommon in patients with S endocardial cushion defects, with the incidence ranging from 3% to 7% [1-4]. The deficiency of the muscular septum and the abnormal displacement of mitral valve result in a long left ventricular outflow tract with predisposition for subaortic stenosis [5-7]. The obstruction can be due to the presence of discrete fibromuscular tissue, tunnel outflow, or malattachment of endocardial cushion tissues to the outflow tract. Serious subaortic stenosis may already be present at the time of repair of the endocardial cushion defect or may develop later. The diagnosis may be missed preoperatively as the gradient across the outflow tract can be masked by the atrioventricular septa1 defect, or the diagnosis may be delayed due to inadequate follow-up [, 81. The presence of inherently narrow left ventricular outflow tract and of complex endocardial cushion abnormalities contributing to the obstruction poses a surgical dilemma. Simple resection is often inadequate and carries a high mortality rate [9, 101. This report is a review of our experience in the surgical management of subaortic stenosis associated with endocardial cushion defect. Presented at the Twenty-seventh Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Feb 18-0, Address reprint requests to Dr DeLeon, Loyola University Medical Center, 160 First Ave, Maywood, IL Material and Methods Patients Over a 15-year period ending in 1989, 1 patients with endocardial cushion defect had subaortic stenosis requiring operative intervention. The clinical profile is outlined in Table 1. Mean age was 7 t 6 years and mean peak systolic subaortic gradient was 60 t 5 mm Hg. Only 3 patients had Down s syndrome. Three patients had coarctation of the aorta. Two of these patients underwent coarctation repair with additional pulmonary artery banding followed by subclavian pulmonary artery shunt in 1 (patient ) before repair of the endocardial cushion defect. One patient (patient 8) also had partial anomalous pulmonary venous drainage involving the right upper lobe that was repaired along with the coarctation at the time of endocardial cushion defect correction. In patients (patients 1 and ), the subaortic stenosis from discrete fibromuscular tissues was recognized before repair of the endocardial cushion defect. In the remaining 10 patients, the subaortic stenosis was diagnosed to 14 years (mean, 6.3 t 5 years) after the repair of the endocardial cushion defect. The obstruction was due to discrete fibromuscular tissues in 5, mitral valve malattachment in 3, and tunnel outflow in (Figs 1, ). In 3 patients left ventricular angiography before repair of the endocardial cushion defect was suggestive of left ventricular outflow tract obstruction although no pressure gradient was demonstrated by The Society of Thoracic Surgeons OO /91/$3.50

2 Ann Thorac Surg 1991;5:107&83 DELEON ET AL 1077 Table 1. Clinical Profile Down s Patient Age Syndrome1 Type of Initial Repair Gradient No. (Y) Coarctation ECD to SAS (y) Type of SAS (mm Hg) D DIC... C D C a Time from repair of ECD to diagnosis of SAS. C = coarctation; D = Down s syndrome; ECD = endocardial cushion defect; SAS = subaortic stenosis Tunnel Mitral valve Mitral valve Tunnel Mitral valve Operative Procedures Two patients (patients 1 and ) underwent resection of discrete fibromuscular tissues through a right atrial approach by retraction of the anterior bridging leaflet at the time of complete atrioventricular canal repair. Ample patch was used to close the ventricular septal defect underneath the anterior bridging leaflet and to enlarge the left ventricular outflow tract. The second patient also had pulmonary artery debanding and closure of a subclavianpulmonary artery shunt. In 5 patients (patients 3 through 7) with discrete fibromuscular narrowing diagnosed after the endocardial cushion repair, generous resection of muscle along with fibrous tissues was performed through an aortotomy. In patients (patients 8 and 9) in whom the obstruction was caused by malattached mitral valve and tunnel outflow, a left ventricular apicoaortic Dacron conduit with porcine valve was inserted (Fig 3) (11, 11. The conduit exited through the pericardium, coursed upward medial to the left lung, reentered the pericardium, and was anastomosed to the proximal ascending aorta. In other patients (patients 10 and 11) in whom the obstruction was also due to malattached mitral valve and tunnel outflow, the modified Konno procedure was performed (Fig 4) [13, 141. After bicaval and ascending aorta cannulation for cardiopulmonary bypass, the right ventricular outflow was opened. Through an aortotomy the subaortic area was assessed, the aortic valve protected, and a conal incision made just inferior to the pulmonary valve leaving a rim of tissue for subsequent placement of sutures. The conal septum was then enlarged using a polytetrafluoroethylene patch with pledgeted sutures. The infundibulotomy was with a polytetrafluoroethylene patch, and the aortotomy closed primarily. In the last patient with Severe subaortic stenosis diagnosed 11 years after an ostium primum repair, resection Fig 1. Left Ventricular view of a heart specimen with atrioventricular septal defect showing anterior displacement of the anterior leaflet of the initral valve causing outflow narrowing. Accessory mitral valve (arrow) and chorda (arrows) contributing to obstruction are shown. (Ao = aorta.)

3 1078 DELEON ET AL Ann Thorac Surg 1991;5:107&83 A C Fig. (A) Left ventriculograrii (patient 4) showin<y discrete sirliaortic stenosis. (B) Left ventriculograni (patient 11) showing tunnel outflow. The anterior displaceinent of the rtiitral valve probably is contributory. (C) Left uentviculopmi (patient 10) shouiing widattached rnitral zialoc causing outflo'iu obstruction. B Fig 3. Apicoaortic vdved conduit inserted in patients. The conduit coursed hetuieen the pericardiuni and left lung.

4 Ann Thorac Surg 1991;5: DELEON ET AL 1079 Fig 4. Modified Konno procedure ( patients). The incision is marked in the infundibulum, ascending aorta, and conal septum. Aortotomy is done for assessment of the outflow and protection of the aortic value. The outflow is enlarged with a polytetrafluoroethylene patch. The infundibulotomy is also closed with a patch.

5 1080 DELEON ET AL Ann Thorac Surg 1991 :5:107&83 and myotomy was performed. Two years later, cardiac catheterization revealed a subaortic gradient of 100 mm Hg secondary to the malattached mitral valve. Through a right atrial approach, the atrial septum was opened. The malattached mitral valve was detached from the top of the ventricular septum and a ventricular septal defect was created (Fig 5). A pericardial patch was then used to close the ventricular septal defect, enlarging the left ventricular outflow tract. The mitral valve was reattached to the upper edge of the pericardial patch [8, 151. Results Early The postoperative course in all patients was uneventful. Late One patient died of Staphylococcus aureus sepsis years after insertion of the apicoaortic conduit. Postmortem examination showed infection of the conduit and abscesses in the brain, kidney, and spleen. Unexplained fever and severe mitral insufficiency developed months postoperatively in the patient who underwent lifting of the malattached mitral valve. Although no organism was isolated, infective endocarditis was suspected and treated. The mitral valve was replaced with a porcine valve and the patient did well with no evidence of subaortic stenosis. Replacement of the porcine valve with a metallic prosthetic valve was required 7 years later. Postoperative echocardiography performed 1 month to 10 years (mean, 3? 3 years) in 9 patients showed a subaortic gradient ranging from 0 to 36 mm Hg (mean, 10.5? 14 mm Hg). In the patients who had a modified Konno procedure, the subaortic gradient was 0 and 0 mm Hg, respectively, with the gradient localized across a mildly hypoplastic aortic valve annulus (Fig 6). Ten patients who have been followed up for up to 15 years (mean, 5 & 4 years) are clinically well, and none have shown serious subaortic stenosis. One patient (patient 5) was lost to follow-up. Comment The presence of an elongated left ventricular outflow tract in patients with endocardial cushion defects raises the possibility of potential or actual subaortic stenosis, which remains a major surgical problem [5-71. Piccoli and colleagues [3] found unequivocal subaortic stenosis in 8 of 114 heart specimens (7%) with atrioventricular septal defects. DeBiase and colleagues [l] reported subaortic stenosis in 4 of 90 consecutive patients (5%) with atrioventricular canal. Taylor and Somerville [] found fixed subaortic stenosis developing in 3 of 99 patients (3%) after repair of ostium primum defects. Subaortic stenosis can be due to discrete fibromuscular tissues in the left ventricular outflow tract or to actual tunnel outflow narrowing. In addition, several endocardial cushion tissue abnormalities can occur that may be primary or contributory factors to the obstruction. These include anterior displacement of the mitral valve, malattachment of chordae and papillary muscle, presence of Fig 5. (A) Lifting of malattached mitral valve using a right atrial approach and incision through atrial septum (1 patient). The area is marked for detaching the anterior mitral leaflet from the left ventricular outflow tract, creating a ventricular septal defect. IB) The ventricular septal defect is closed with a putch enlarging the left ventricular outflow tract, and the anterior mitral leaflet is reuttached to the top of the putch. accessory mitral valve tissue, or presence of accessory tricuspid tissue protruding through a ventricular septal defect into the outflow tract [ The diagnosis of subaortic stenosis based on conventional pressure data may be difficult during cardiac catheterization because of the potential masking effect of the left to right shunt on the gradient across the left ventricular outflow tract [8]. Our patients who had simultaneous repair of subaortic stenosis and canal had low preoperative outflow gradients. Lappen and colleagues [8] and Taylor and Somerville [] claim that persistence or exaggeration during systole of the characteristic angiographic diastolic "gooseneck" deformity is diagnostic of potential or actual subaortic stenosis in endocardial cushion defect.

6 Ann Thorac Surg 1991;5:107&83 DELEON ET AL 1081 Fig 6. Echocardiograms of the patients who had the modified Konno procedure. Preoperative pictures (left) show the narrowed left ventricular outflow tract, which became quite wide after the procedure (right). (Ao = aorta; LA = left atrium; LV = left ventricle; RV = right ventricle.) A B In patients who had previous repair of atrioventricular septa1 defects, the possibility of subsequent subaortic stenosis cannot be ignored. Closure of the cleft of the anterior leaflet of the mitral valve may contribute to the narrowing of the left ventricular outflow tract [1]. Early detection of subaortic stenosis requires a high index of suspicion and follow-up endocardiographic examination looking for systolic fluttering and early closure of the aortic valve, narrowing of the left ventricular outflow tract, left ventricular hypertrophy, and increased systolic flow velocity across the outflow tract []. The appropriate surgical technique for relieving the subaortic stenosis chiefly depends on the anatomy of the outflow tract. Despite the important data provided by echocardiography and angiocardiography, the intraoperative findings ultimately determine the surgical technique that is needed. When the left ventricular outflow tract is not too narrow, relief of the obstruction by resection of discrete fibromuscular tissues may suffice, provided that generous myectomy is also performed. However, if the left ventricular outflow tract is inherently narrowed, resection of fibromuscular tissues alone may be ineffective

7 108 DELEON ET AL Ann Thorac Surg 1991:5: and carries a high mortality risk as well as recurrence rate [9, 10, 11. The increased turbulence across a narrowed subaortic area stimulates connective tissue proliferation, increased deposition of fibrous tissue, and further progression of obstruction in a vicious cycle []. When subaortic stenosis without a substantial outflow gradient is suspected before repair of the endocardial cushion defect, exploration of the outflow tract and resection of fibromuscular tissues along with the repair of the atrioventricular septal defect is a reasonable approach. In patients in whom a serious outflow obstruction as measured by serial angiography or echocardiography has developed after atrioventricular septal defect repair, intraoperative exploration should help in deciding between simple fibromuscular resection alone or a more aggressive surgical approach such as insertion of an apicoaortic conduit, modified Konno procedure, or lifting of malattached mitral valve. Where the subaortic stenosis is recurrent, a more aggressive surgical approach is indicated [1. Although the modified Konno procedure has been reported in only a few cases [14, 31, it is preferable over insertion of an apicoaortic valved conduit as the native aortic valve is preserved. Long-term anticoagulation and possible replacement of valve prosthesis are avoided with the modified Konno procedure. Lifting of the malattached anterior mitral leaflet from the left ventricular outflow tract is an alternative. Experience with the technique, however, is quite limited and there is a risk of mitral insufficiency [8, 151. References 1. DeBiase L, DiCiommo V, Ballerini L, Bevilacqua M, Marcelletti C, Marino B. Prevalence of left-sided obstructive lesions in patients with atrioventricular canal without Down s syndrome. J Thorac Cardiovasc Surg 1986;91: Taylor NC, Somerville J. Fixed subaortic stenosis after repair of ostium primum defects. Br Heart J 1981;45: Piccoli GP, Wilkinson JL, Macartney FJ, Gerlis LM, Anderson RH. Left-sided obstructive lesions in atrioventricular septal defects. J Thorac Cardiovasc Surg 198;83: Gow RM, Freedom RM, Williams WG, Trusler GA, Rowe RD. Coarctation of the aorta or subaortic stenosis with atrioventricular septal defect. Am J Cardiology 1984;53: Ebels T, Ho SY, Anderson RH, Meijboom EJ, Eijgelaar A. The surgical anatomy of the left ventricular outflow tract in atrioventricular septal defect. Ann Thorac Surg 1986;41: Wright JS, Newman DC. and intermediate atrioventricular canal in infants less than a year old: observations of anatomical and pathological variants in left ventricular outflow tract. Ann Thorac Surg 198;: McGrath LB, Kirklin JW, Soto B, Bargeron LM Jr. Secondary left atrioventricular valve replacement in atrioventricular septal (AV canal) defect: a method to avoid left ventricular outflow tract obstruction. J Thorac Cardiovasc Surg 1985;89: Lappen RS, Muster AJ, Idriss FS, et al. Masked subaortic stenosis in ostium primum atrial septal defect: recognition and treatment. Am J Cardiol 1983;5: Heydarian M, Griffith BP, Zuberbuhler JR. atrioventricular canal associated with discrete subaortic stenosis. Am Heart J 1986;109: Ben-Shachar G, Moller JH, Castaneda-Zuniga W, Edwards JE. Signs of membranous subaortic stenosis appearing after correction of persistent common atrioventricular canal. Am J Cardiol 1981;48: Ergin MA, Cooper R, LaCorte M, Golinko R, Griepp R. Experience with left ventricular apicoaortic conduits for complicated left ventricular outflow obstruction in children and young adults. Ann Thorac Surg 1981;3: Norman JC, Nihill MR, Cooley DA. Valved apico-aortic composite conduits for left ventricular outflow tract obstructions. Am J Cardiol 1980;5: Konno S, Imai Y, Iida Y, Nakajima M, Tatsuno K. A new method for prosthetic valve replacement in congenital aortic stenosis associated with hypoplasia of the aortic valve ring. J Thorac Cardiovasc Surg 1975;70: Cooley DA, Garrett JR. Septoplasty for left ventricular outflow obstruction without aortic valve replacement: a new technique. Ann Thorac Surg 1986;4:44M. 15. Chang Cl, Becker AE. Surgical anatomy of left ventricular outflow tract obstruction in complete atrioventricular septal defect. J Thorac Cardiovasc Surg 1987;94: Sellers RD, Lillehei CW, Edwards JE. Subaortic stenosis caused by anomalies of the atrioventricular valves. J Thorac Cardiovasc Surg 1964;48: Hatem J, Sade RM, Taylor A, Usher BW, Upshur JK. Supernumerary mitral valve producing subaortic stenosis. Chest 1981;79: Cooperberg P, Hazel1 S, Ashmore PG. Parachute accessary anterior mitral valve leaflet causing left ventricular outflow tract obstruction. Circulation 1976;53:90& Bjork VO, Hultquist G, Lodin H. Subaortic stenosis produced by an abnormally placed anterior mitral leaflet. J Thorac Cardiovasc Surg 1961;41: Nanton MA, Belcourt CL, Gillis DA, Krause VW, Roy DL. Left ventricular outflow tract obstruction owing to accessory endocardial cushion tissue. J Thorac Cardiovasc Surg 1979; 78: Spanos PK, Fiddler GI, Mair DD, McGoon DC. Repair of atrioventricular canal associated with membranous subaortic stenosis. Mayo Clin Proc 1977;5:114.. DeLeon SY, Ilbawi MN, Arcilla RA, et al. Transatrial relief of diffuse subaortic stenosis after ventricular septal defect closure. Ann Thorac Surg 1990;49: Kirklin JW, Barratt-Boyes BG. Congenital aortic stenosis. In: Kirklin JW, Barratt-Boyes BG, eds. Cardiac surgery. New York: Wiley, 1986: DISCUSSION DR JOSEPH J. AMATO (New Hyde Park, NY): I congratulate Dr DeLeon and associates on the superb management of a difficult situation. I have, however, two questions for them. First, because only two of these 1 patients with endocardial cushion defects were diagnosed as having severe subaortic stenosis, is it possible that some technical problem with your use of the two-patch technique could have been the cause of the development of fibromuscular tissue or valve malattachment in the other 10 patients to 14 years later? Second, can you clarify your statement regarding accessory

8 Ann Thorac Surg 1991;5: DELEON ET AL 1083 mitral valve tissue? Was the tissue found actual valvular tissue or excessive fibrous tissue developed from obstructive flow? DR DELEON: The majority of our patients had an ostium primum defect, although 5 also had a complete canal. What we have been warning all along is that when you repair a complete canal you should use an ample patch and not just suture the anterior bridging leaflet down to the ventricular septum, converting it to ostium primum and thereafter closing the atrial septal defect. We use a two-patch technique so that we do not interfere much with the mitral or tricuspid valve. DR ADNAN COBANOGLU (Portland, OR): I was surprised at the number of discrete subaortic stenosis cases that you encountered. I would expect almost all of these cases to have a structural basis to them or to develop in the manner that you describe with fibrous tissue deposition in the subaortic region over a period of years. I think that even after a local myectomy, the underlying structural abnormality, which you very nicely indicated, is still there. I wonder if at 10 years of follow-up you will not see some of these patients come back for a third operation because of recurrence of a similar pathology in exactly the same area in the left ventricular outflow tract. DR DELEON: That is an excellent question. The incidence I quoted of 3% to 7% was mostly from pathology specimens. I think the lack of awareness of this defect resulted in very few cases operated on in the past, but now, with increased awareness, we should be able to recognize this problem more often. As I indicated in the presentation, the outflow tract is quite long in patients with endocardia1 cushion defect, which leads to turbulence in the area with increased tissue deposition. If you are not aware of this phenomenon, you may just resect what you might think are discrete tissues causing the obstruction, not knowing that the inherent problem is a long, narrowed outflow tract. In the past, this has led to increased recurrence and mortality rates. I think I failed to answer one of Dr Amato s questions. Echocardiography and angiography of these patients would suggest the presence of tunnel outflow or malattachment of the mitral valve. The ultimate decision as to what to do, however, is in the intraoperative exploration. With recognition of an obstruction at the time of the initial presentation, resection of the obstructing tissues along with the repair of the canal should be adequate, unless there is no question that there is tunnel outflow or malattached mitral valve. In some patients in whom subaortic stenosis develops after canal repair, it will be obvious that there is tunnel outflow or malattached mitral valve. However, there will be certain patients who will fall into a gray zone. In these patients, do you do resection or do you go ahead with an invasive approach? In these patients, you may be justified to do resection with generous myectomy. In the event of recurrence, I think you are justified to go ahead with a more aggressive approach such as the modified Konno procedure. DR COBANOGLU: Intraoperatively in these cases you have control as far as the diameter of the left ventricular outflow tract is concerned if you are correcting a complete atrioventricular canal defect, because you can move the ventricular septal defect patch toward the right side of the septum and have the outflow area larger than you would by suturing onto the crest. I think the problem is in the partial canal defects, where the leaflet is already tethered and attached to the crest of the ventricular septum; at least in our experience, those have been the more difficult cases to deal with when we are aware that there is some preoperative left ventricular outflow obstruction at the time of total correction.