Shone’s Anomaly: Operative Results and Late Outcome Steven F. Bolling, MD, Mark D. Iannettoni, MD, Macdonald Dick 11, MD, Amnon Rosenthal, MD, and Edward L. Bove, MD Sections of Thoracic Surgery and Pediatric Cardiology, The University of Michigan Medical School, Ann Arbor, Michigan
Shone‘s anomaly, a congenital cardiac malformation, consists of multiple levels of left heart obstruction including supravalvar mitral ring, parachute mitral valve, subaortic stenosis, and coarctation. The prognosis for patients with Shone’s anomaly is poor. To assess operative results and late outcome, we reviewed the records of 30 consecutive patients seen with Shone’s anomaly at our institution between 1966 and 1989. Anatomical diagnoses in these patients were supravalvar mitral ring (22 patients), mitral valve abnormalities including parachute mitral valve, fused chordae, or single papillary muscle (26 patients), subaortic gradients (26 patients), and coarctation (29 patients). Nineteen patients had all four lesions. Other common defects were bicuspid aortic valve (19 patients) and ventricular septal defect (20). Two patients were treated medically. The other 28 patients required 84 operative procedures with 18 patients undergoing more than one procedure. Operations included coarctation repair (28 patients), mitral valve repair or replacement (111, ventricular septal defect closure (8),
subaortic resection (8), and complex left ventricular outflow tract reconstruction or bypass (4). Age at first operation ranged from 7 days to 7 years (median age, 3 months). There were no operative deaths at the first operation. However, mortality rose to 24% (4117) after the second operation. All operative deaths were secondary to severe mitral valve disease. The survivors have been followed from 1 to 16 years (mean follow-up, 6 f 1 years). There were no late or sudden deaths. Morbidity has included stroke (l), gastrointestinal bleeding (21, permanent heart block ( l ) , and persistent congestive heart failure (6). We conclude that although patients with Shone’s anomaly are seen with a wide spectrum of major anatomical and hemodynamic abnormalities, a satisfactory long-term outcome is possible for most. Operative mortality is adversely affected by the severity of mitral valve disease and the need for multiple operative procedures.
S
Material and Methods
ince the first description of the developmental complex of parachute mitral valve, supravalvar ring of the left atrium, subaortic stenosis, and coarctation of the aorta by Shone and associates [l]in 1963, little clinical experience has accumulated about the operative therapy or late follow-up of patients with Shone’s anomaly. Cardiologists and cardiac surgeons have reported patients with Shone‘s complex to have an unsatisfactory operative outcome and a poor long-term prognosis. Although patients with Shone’s anomaly tend to be grouped together in terms of therapeutic approach, these patients actually demonstrate a wide spectrum of developmental malformations. We postulated that if the operative approach to these patients is tailored and individualized, the long-term outcome might be better. To assess the operative results and the late outcome with such an approach, we retrospectively reviewed the data from 30 consecutive patients with the diagnosis of the developmental complex of Shone’s anomaly.
Presented at the Twenty-fifth Anniversary Meeting of The Society of Thoracic Surgeons, Baltimore, MD, Sep 11-13, 1989. Address reprint requests to Dr Bolling, University of Michigan Hospitals, 1500 E Medical Center Dr, 2120 TC/0344, Ann Arbor, MI 48109.
0 1990 by
The Society of Thoracic Surgeons
(Ann Thorac Surg 1990;49:887-93)
The records of 30 consecutive patients seen at The University of Michigan between 1966 and 1989 were examined for the severity of Shone‘s anomaly, the operative approach, and the eventual late outcome. Reviewed at this time were operative notes, cardiac catheterization reports, inpatient notes, and referring letters. The patients were followed up by a combination of review of outpatient clinic notes and consultation with each patient’s private physician. There were 16 female and 14 male patients. Height and weight at the first operation were 74 2 9 cm and 8.0 t 1.7 kg, respectively. Most patients fell between the fifth and tenth percentile for height and weight at the time of first presentation. Seventeen of the patients were first seen at The University of Michigan in frank congestive heart failure (CHF) necessitating hospitalization. Sixteen of the 17 were on a regimen of digitalis and a diuretic at the time of admission. The condition of 2 additional patients was well controlled with digitalis and a diuretic without evidence of CHF at the time of first admission. Five patients had cardiomegaly on the chest roentgenogram without physical signs of CHF. Four patients had an unremarkable chest roentgenogram and only a murmur on physical 0003-4975/90/$3.50
888
BOLLING ET AL SHONE'S ANOMALY
Ann Thorac Surg 1990;49:887-93
examination, and 1patient with a normal chest roentgenogram was cyanotic on admission. Twelve of the 18 patients who underwent a second operative repair were in CHF. Nine of these 12 patients were receiving digoxin, a diuretic, a vasodilator, or a combination of these before admission. All patients had systolic ejection murmurs greater than 3/6. Four associated syndromes were detected in 3 patients: adrenogenital syndrome, Turner's syndrome, phocomelia, and "floppy baby syndrome." Normal sinus rhythm was demonstrated in 28 patients, and the mean axis was 93 k 8 degrees. One patient was noted to have first-degree heart block. Seven patients had electrocardiographic evidence of left ventricular hypertrophy, 7 patients showed right ventricular hypertrophy, and 9 patients demonstrated biventricular hypertrophy.
Catheterization Data Right and left heart catheterization data obtained in 26 children were compared. Two patients could not undergo left heart catheterization because of technical difficulties, and the records of 2 patients were incomplete. The pressures (mean ? standard error of the mean, systolic/diastolic) for each cardiac chamber were as follows: right atrium, 5 k 1 mm Hg; right ventricle, 62 k 7/7 k 1 mm Hg; pulmonary artery, 60 5/24 & 3 mm Hg; left ventricle, 135 ? 8/13 k 2 mm Hg; ascending aorta, 114 k 7/67 ? 4 mm Hg; and descending aorta, 86 ? 5/60 f 4 mm Hg. Systolic gradients at the area of subaortic obstruction ranged from 10 to 100 mm Hg (mean gradient, 21 mm Hg) and coarctation gradients ranged from 12 to 75 mm Hg (mean gradient, 28 mm Hg). The oxygen saturation was as follows: right atrium, 72% k 1%;right ventricle, 75% ? 1%;pulmonary artery, 76% 5 2%; left ventricle, 94% k 1%;and aorta, 89% k 3%.
*
Anatomy Anatomical evaluation was done in all patients by echocardiography, cardiac catheterization, or both, and then confirmed at operation. The aspects of Shone's anatomy for each patient are shown in Table 1. A diagrammatic representation of the four obstructive lesions in Shone's anomaly is shown in Figure 1. Coarctation of the aorta, found in 29% of the 30 patients, was the most common defect. Subaortic stenosis due to a fibromuscular hypertrophy (25 patients) or a subvalvar membrane (1 patient) was also observed. Abnormalities of the mitral valve were seen in 26 patients and included mitral stenosis, fused chordae, single papillary muscle, and parachute mitral valve apparatus. A supravalvar mitral ring was identified in 22 patients. Nineteen patients had all four components of Shone's anomaly. These were the patients with so-called pure Shone's anomaly. Other associated defects included ventricular septa1 defect (VSD) (20 patients), patent ductus arteriosus (12), and bicuspid aortic valve (19). There were 2 patients with additional left-sided malformations noted at catheterization. One patient had a hypoplastic left ventricle and the other, a severely hypoplastic aortic arch. Other valvar abnormalities included
moderate tricuspid regurgitation (3 patients) and mild pulmonary stenosis (2).
Operative Data Of the 30 patients with Shone's anomaly, 28 underwent operative intervention. The operative procedures are summarized in Table 2. Two patients were treated medically. One patient, an 8-year-old child, was considered to have only minimal hemodynamic abnormalities, despite the presence of all four lesions of Shone's anomaly. The other patient, a 3-month-old infant, was considered to have defects that could not be corrected by operation. Age at the first operation ranged between 7 days and 7 years (mean age, 11 months; median age, 3 months). Nineteen patients had the first operation before 1year of age. There were 42 procedures performed at the first operation in 28 patients, an average of 1.5 2 0.2 per patient. The most common procedure was coarctation repair (23 patients), followed by ligation of a patent ductus arteriosus (10). Other procedures at first operation included mitral valvotomy (3), aortic valvotomy (2), and atrial septostomy, pulmonary artery banding, subaortic resection, and a Blalock-Taussig shunt (1 each). Aortic cross-clamp time was 27 ? 4 minutes for procedures requiring cardiopulmonary bypass. The second operative procedure, necessary in 18 patients, was performed at a mean age of 46 months (range, 1 month to 8 years). The average number of procedures was 1.6 k 0.4 (28 procedures in 17 patients). These operations included VSD repair (7 patients), reoperation for coarctation (4), aortic valvotomy (3), mitral valvotomy (2), subaortic resection of a muscular band (n = 2), and left ventricular outflow tract myotomy, excision of a subaortic membrane, placement of a Hancock bioprosthesis in the mitral position, descending aorta-main pulmonary artery shunt, apicoaortic conduit, pulmonary artery banding, Konno aortoventriculoplasty with St. Jude aortic valve replacement and St. Jude mitral valve replacement, Omniscience mitral valve replacement, resection of a right ventricular muscle band, placement of a permanent pacemaker, and tricuspid valvotomy (1 each). A third operation was performed in 6 patients at a mean age of 52 months (range, 6 months to 7 years). These procedures included right ventricular outflow tract enlargement (2 patients), reoperation for coarctation (2), replacement of an apicoaortic conduit, and placement of a left subclavian artery-descending aortic shunt. Three patients required a fourth operation at 2, 7, and 9 years of age. These procedures included placement of a subclavian artery-descending aortic shunt, pacemaker placement, and reoperation for coarctation. One patient required a fifth operation, a reoperation for recurrent subaortic stenosis, previously resected when the patient was 2 years of age. Six patients underwent reoperation for recurrent coarctation. Two of them required shunting from the left subclavian artery to the descending aorta because of inadequate flow from the left ventricle to the descending aorta owing to restriction at the level of the coarctation. No patient had more than one surgical revision of the coarctation, and 2 patients had balloon angioplasty for
BOLLINGET AL SHONE'S ANOMALY
Ann Thorac Surg 1990;49:887-93
889
Table 1. Summary of Anatomical Data for 30 Patients With Shone's Anomaly Lesions in Shone's Anomaly Patient No.
Left Atrial Ring
Mitral Valve Abnormality
Subaortic Obstruction
Coarctation
All Four
1 2 3 4
No No Yes Yes
Stenosis Normal Fused chordae Stenosis
No Yes Yes No
Yes Yes Yes Yes
No No Yes No
5 6
Yes Yes
Stenosis All to 1 papillary
No Yes
Yes Yes
No Yes
7 8 9 10 11 12 13 14 15 16 17 18
No Yes Yes Yes No No Yes No Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes
Fused chordae Stenosis Stenosis Fused chordae Normal Normal Parachute mitral Stenosis/regurgitation Stenosis Parachute mitral All to 1 papillary Stenosis Stenosis Stenosis Stenosis Fused chordae All to 1 papillary All to 1 papillary Fused chordae All to 1 papillary Parachute mitral Stenosis Normal Stenosis
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes
No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
No Yes Yes Yes No No Yes No Yes No Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes No Yes
19
20 21 22 23 24 25 26 27 28 29 30
ASD = atrial septa1 defect; septa1 defect.
Bi Ao = bicuspid aortic valve;
FDA = patent ductus arteriosus;
this problem. Five patients with coarctation underwent attempts at balloon angioplasty that were considered unsuccessful. Two patients underwent reoperation for recurrent subaortic obstruction that had previously been resected, and 1 patient underwent mitral valve replacement 1 month after a failed mitral commissurotomy. In total, 84 operative procedures were performed in 28 patients with an average of 2.8 separate procedures per patient occurring at an average of 2.4 operations per patient.
Results Mortality There were no operative deaths associated with first operations. All four deaths occurred in the hospital after
Other Anomalies VSD, PDA PDA, PFO, Bi Ao VSD, Bi Ao VSD, PDA, PFO, ASD, hypoplastic arch, Bi Ao None VSD, PDA, PFO, ASD, hypoplastic left ventricle Bi Ao VSD, Bi Ao VSD, Bi Ao VSD, PDA, Bi Ao VSD, Bi Ao VSD, Bi Ao PDA, Bi Ao VSD, PDA, PFO, Bi Ao PDA None VSD, Bi Ao VSD, PDA, PFO, ASD VSD, PDA None PFO, Bi Ao VSD, Bi Ao Bi Ao VSD, PDA, PFO VSD, PDA VSD Bi Ao VSD, Bi Ao VSD, Bi Ao VSD, Bi Ao
PFO = patent foramen ovale;
VSD = ventricular
the second operation (4/17).The first operative death was attributed to low cardiac output after Hancock bioprosthetic valve placement in the mitral position and VSD closure in a child who remained in severe CHF 1 month after a coarctation repair and PDA ligation. This child was 3 months old at the time of death. The second postoperative death occurred in a 3year-old child after coarctation repair and a systemicpulmonary artery shunting procedure, done 2 weeks after an atrial septostomy and mitral valvotomy. The third postoperative death was in a 4-year-old girl who underwent a Konno aortoventriculoplasty with St. Jude prosthetic valve placement in both the aortic and mitral positions. This patient had had a previous flap angioplasty repair of coarctation immediately after birth and remained in severe CHF despite medications. Postopera-
890
BOLLINGET AL SHONE'S ANOMALY
Ann Thorac Surg 1990;49:887-93
Fig 1. The four obstructive anomalies forming Shone's anomaly: supravalvar ring of the left atrium, parachute mitral value, subaortic stenosis, and coarctation of the aorta, in that order according to the direction of blood flow. (Reprinted from [I] by permission of the American lournal of Cardiology.)
tive death was attributed to low cardiac output from right coronary artery occlusion. The fourth patient who died had phocomelia and floppy baby syndrome and was returned to the hospital at age 23 months for a second operation. The second operation included closure of an atrial septa1 defect and unbanding of the pulmonary artery. At that time, the patient had a restrictive VSD with suprasystemic right ventricular pressures and severe desaturation. Aplasia of the bone marrow and severe thrombocytopenia developed and ultimately led to pulmonary hemorrhage and death. Three of 4 patients who died had all four anatomical anomalies of Shone's. There have been no late deaths. The aortic cross-clamp time for the first operation requiring cardiopulmonary bypass was not significantly different between survivors and nonsurvivors (23 & 8 minutes versus 40 5 minutes, respectively). When catheterization data of operative survivors were compared with those of nonsurvivors, the only significant differences were for pulmonary artery systolic and diastolic pressures. The pulmonary artery systolic pressures 14 mm Hg and were 62 & 2 mm Hg versus 102 pulmonary artery diastolic pressures, 22 3 mm Hg versus 39 f 9 mm Hg for survivors and nonsurvivors, respectively (both, p < 0.05).
*
*
*
Morbidity There was one serious postoperative complication resulting in a prolonged hospital course after the first operation. It was a sternal wound infection, which resolved with
intravenous antibiotics and secondary closure. No organisms were ever identified. In addition, 1 patient was transferred to our institution with the symptoms of peritonitis and guaiac-positive stools. This patient had all four abnormalities of Shone's and was in florid CHF at the time of presentation. Necrotizing enterocolitis developed after the cardiac operation, and the patient required prolonged hyperalimentation and hospitalization. No abdominal procedure was necessary as a result of this complication. The incidence of major morbidity increased with each succeeding operation. After the second operation, 1 patient sustained a series of postoperative complications, which included a posterior myocardial infarction, reexploration for bleeding, an upper gastrointestinal bleed, stroke, and deficiency of coagulation factors I1 and V. This patient had replacement of the apicoaortic conduit and enlargement of the right ventricular outflow tract and is now doing well 19 months after operation. A series of conduction abnormalities due to complete heart block were noted in 1 patient after operation and resulted in placement of a permanent pacemaker. One patient each had right bundle-branch block, right ventricular conduction delay, and left bundle-branch block, but these conditions did not necessitate placement of a permanent pacemaker.
Follow-up Data Of the 30 patients identified with Shone's anomaly, 4 patients died and no follow-up was available for 1patient. Of the 2 patients treated medically, 1 returned to Japan after the diagnosis of Shone's anomaly was made, and she was lost to follow-up. This young Oriental girl had adrenogenital syndrome as well as Shone's anomaly. The other medically treated patient had Turner's syndrome as well as mild Shone's anomaly, and is now working as a secretary. Of the 28 patients who underwent surgical intervention, 24 were eligible for follow-up after their last operative procedure. The average length of follow-up is 5.7 & 1.1 years after the most recent operation. The latest New York Heart Association activity level was class I for 82%, class I1 for 12%, and class I11 for 6%. Seven patients have persistent cardiomegaly on chest roentgenograms. Six patients still require medical therapy for control of persistent episodes of CHF. At the most recent postoperative catheterization, the mean follow-up gradient through the area of repaired aortic coarctation was 16 2 4 mm Hg (7 patients) and across the aortic outflow tract, 33 10 mm Hg (5 patients). Two patients still have apicoaortic conduits in place. Only 4 patients were reported to have a residual gradient across the mitral valve (mean gradient, 5 1 mm Hg). Other reported valvar abnormalities at follow-up were mild mitral regurgitation in 3 patients and moderate tricuspid regurgitation in 1 patient. Although these patients with Shone's anomaly remain quite small physically, success in school was at least average in this group based on follow-up reports. Furthermore, only 3 of the patients in the follow-up group were
*
*
BOLLING ET AL SHONE‘S ANOMALY
Ann Thorac Surg 1990;49:887-93
891
Table 2. Summary of Operative Procedures From Most Recent to lnitial Patient No. 1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
30
Operative Repairs and Procedures
1: Redo CoA repair with Dacron; 2: VSD closure; 3: CoA repair and PDA ligation 1: CoA repair with Gore-Tex patch and PDA ligation 1: Mitral valvotomy; 2: CoA balloon angioplasty for restenosis; 3: CoA repair 1: Redo CoA repair with Dacron; 2: CoA repair 1: CoA repair with LSCA patch No repair 1: Open aortic and mitral commissurotomies 1: CoA repair (end to end) 1: Subaortic resection of muscle band, VSD closure; 2: CoA repair 1: Aortic valvotomy, LV outflow tract myomectomy, VSD closure; 2: subaortic membrane excision; 3: FDA ligation, CoA repair 1: Resection of subaortic restenosis; 2: redo CoA repair; 3: resection and patch of RV outflow tract obstruction; 4: VSD closure; 5: PDA ligation, CoA repair 1: CoA repair and PDA ligation 1: CoA repair and PDA ligation 1: Hancock bioprosthetic replacement of mitral valve, VSD closure; 2: CoA repair, PDA ligation 1: CoA repair, Potts shunt; 2: atrial septostomy, mitral valvotomy No repair 1: CoA repair with LSCA 1:PA banding, CoA repair; 2: right Blalock-Taussig shunt, atrial septostomy 1: VSD closure; 2: CoA repair with LSCA, PDA ligation 1: RV outflow tract enlargement, replacement of apicoaortic conduit; 2: LV apex-aorta conduit; 3: CoA repair 1: Mitral valvotomy; 2: CoA repair 1: Redo CoA repair with Dacron; 2: CoA repair with LSCA 1: CoA repair with LSCA; 2: aortic commissurotomy, resection of subaortic stenosis 1: CoA repair with Gore-Tex, PDA ligation 1: Konno aortoventriculoplasty with St. Jude replacement of aortic and mitral valves, CoA balloon angioplasty for restenosis; 2: CoA repair with LSCA 1: PA banding, CoA repair 1: CoA repair (end to end) 1: LSCA-descending aortic shunt; 2: CoA repair (end to end); 3: MVR with No. 17 Omniscience; 4: mitral commissurotomy 1: Pacemaker placement for third-degree heart block; 2: resection of subaortic band, resection of RV band, patch closure of VSD; 3: Ao valvotomy, tricuspid valvotomy, PDA ligation; 4: CoA repair with LSCA 1: PDA, CoA repair with LSCA; 2: Closure of VSD
CoA = coarctation of aorta; LV LSCA = left subclavian artery; artery; FDA = patent ductus arteriosus; RV = right ventricular;
=
considered to be ”minimally” handicapped or to have learning disabilities.
Comment More than 25 years have passed since Shone and colleagues [ 11 first described the developmental complex of parachute mitral valve, supravalvar ring of the left atrium, subaortic stenosis, and coarctation of the aorta. In that original article (8 patients), only 2 patients exhibited all four aspects of Shone’s anomaly and could therefore be considered to have pure cases of Shone’s anomaly. In our series, 19 patients were seen with all four aspects of the anomaly and therefore were considered to have pure cases. The other 11 can be regarded as having forme fruste
MVR = mitral valve replacement; left ventricular; VSD = ventricular septa1 defect.
PA = pulmonary
of Shone’s anomaly, most missing a supravalvar mitral ring. In the description of Shone and associates [l],it was noted that mitral valve obstruction appeared to be the most critical problem associated with the anomaly, and mitral commissurotomy alone (performed in 3 of the patients) was not adequate to relieve the mitral valve obstruction. Shone and colleagues quite farsightedly made a hypothetical case for excision of the supravalvar ring and replacement of the mitral valve with a prosthesis as the best method of overcoming the major mitral valvar and subvalvar obstructions. Our present study confirms the belief that severity of the mitral valve obstruction correlates inversely with long-term outcome and that operative mortality in pa-
892
BOLLING ET AL SHONE'S ANOMALY
tients with Shone's anomaly is adversely affected by the degree of mitral valve disease. The patients who demonstrated substantially elevated pulmonary artery pressures were those with the worst mitral obstruction and had, in general, the poorest outcome. These patients also appeared to have severe and progressive left ventricular outflow tract obstruction. Kieffer and associates [2] noted an association between reversed ventricular septal curvature seen on two-dimensional echocardiography with primary pulmonary hypertension and Shone's anomaly. Indeed, our patients with severe mitral obstructive disease as a result of Shone's anomaly did have secondary pulmonary hypertension, which conceivably could have led to increased reversed ventricular septal curvature and exacerbation of subaortic obstruction. Rosenquist [3] described the results of pathological study of 53 heart specimens in which the primary diagnosis was coarctation of the aorta with normally related great vessels. Rosenquist noted a surprisingly high incidence of obstructive mitral valve involvement in these specimens. Furthermore, the mitral valve abnormalities usually involved the subvalvar apparatus and demonstrated a wide spectrum of anomalies from hypoplastic but otherwise normal mitral valves to severe parachute mitral stenosis. In addition, that report [3] noted three abnormalities in the mitral valves of patients with Shone's anomaly, findings subsequently confirmed by others [4]. These abnormalities involved the mitral subvalvar apparatus and included fusion of the papillary muscles, fusion of a papillary muscle to the ventricular wall, and true parachute mitral valve with only one papillary muscle. Varying degrees of all these forms of subvalvar mitral abnormality were noted in our patients (see Table 1). In a clinicopathological conference report demonstrating the case of a patient with Shone's anomaly [ 5 ] , physicians were cautioned to exercise an awareness of the possibility of other left-sided obstructive lesions that could be present in the symptomatic infant with coarctation. The correlation of obstructive left-sided anomalies associated with coarctation of the aorta was confirmed by Becker and co-workers [6] 2 years later. They believed that Shone's anomaly should be excluded in those patients who continue to have signs of CHF after operative repair of the coarctation. This correlates very well with our present group of patients, many of whom demonstrated continued CHF after the first operation, which was most commonly coarctation repair and ligation of patent ductus arteriosus. As recently as 1987 in a report of spontaneous complete atrioventricular conduction block in Shone's anomaly [71, it was stated that the diagnosis of Shone's anomaly is usually made during infancy or early childhood. However, the authors then noted that most patients with Shone's anomaly are dead soon after the diagnosis is made or soon after an attempt at surgical repair. Many cardiac surgeons and cardiologists agree with this dismal outlook, and therefore, the prevailing sentiment is that the prognosis for these patients is poor. However,
Ann Thorac Surg 1990;49:887-93
few data are available to document the outcome of patients with Shone's anomaly. Although there have been isolated case reports [8] describing the operative approach to the different lesions that make up Shone's anomaly, no overall operative strategy has been developed. Our operative approach to these patients includes early repair of coarctation and ligation of patent ductus arteriosus. Repair of other lesions including VSD can generally be postponed. The majority of VSDs in our series were restrictive and closed spontaneously after coarctation repair. The surgical options for the left ventricular outflow tract obstruction must be tailored to the individual patient because of the wide variability in anatomy. When the aortic annulus is adequate and the obstruction is limited to the subaortic area, septoplasty is an effective procedure. If the annulus is hypoplastic, we favor aortoventriculoplasty (Konno procedure) [9, 101 over left ventricular apex-descending aortic shunts. Aortoventriculoplasty has been used for tunnel subaortic obstruction, and the incision in the interventricular septum can easily be taken well into the left ventricular cavity, effectively relieving the obstruction. The Konno procedure allows the aortic valve to remain in the normal anatomical position with normal flow characteristics and permits the placement of adequately large valves, despite the overall small body size of the patient. This procedure has many other advantages compared with apicoaortic conduit procedures, including decreased arrhythmias and preservation of ventricular function [ll]. The operative treatment of the mitral lesions noted in these patients remains difficult because of the highly variable nature of these malformations and the differing degrees of subvalvar involvement. However, valve repair should be considered whenever feasible in these young patients before mitral valve replacement is performed. The frequency of multiple operations in patients with Shone's anomaly (18 of 30 patients undergoing more than one procedure) makes this group a formidable technical challenge. Furthermore, the high risk of subsequent operations, particularly in those patients with mitral valve involvement, can discourage even the most aggressive surgeon. Finally, although our patients have had a satisfactory long-term outcome, 4 of 24 children have a prosthetic valve in place and must undergo further reoperative procedures to replace bioprosthetic valves as they degenerate or suffer the consequences of anticoagulation, which have led to substantial gastrointestinal bleeding in 2 patients. In summary, our report demonstrates that patients with Shone's anomaly represent a wide spectrum of major anatomical and hemodynamic abnormalities. An aggressive surgical approach tailored to each patient's individual anatomy can result in a satisfactory outcome for most of these patients.
References 1. Shone JD, Sellers RD, Anderson RC, Adams P, Lillehei CW, Edwards JE. The developmental complex of "parachute mi-
Ann Thorac Surg 19’30;49887-93
tral valve,” supravalvular ring of left atrium, subaortic stenosis, and coarctation of aorta. Am J Cardiol 1963;11:71&25. 2. Kieffer RW, Hutchins GM, Moore GW, Buckley BH. Reversed septa1 curvature. Association with primary pulmonary hypertension and Shone‘s syndrome. Am J Med 1978; 66:831-5. 3. Rosenquist GC. Congenital mitral valve disease associated with coarctation of the aorta. A spectrum that includes parachute deformity of the mitral valve. Circulation 1974; 49:98593. 4. Tandon R, Moller JH, Edwards JE. Anomalies associated with the parachute mitral valve: a pathologic analysis of 52 cases. Can J Cardiol 1986;2:27%81. 5. Noren GR. Coarctation of aorta as part of Shone syndrome. Minn Med 1968;9:1617-27. 6. Becker AE, Becker MJ, Edwards JE. Anomalies associated
BOLLING ET AL SHONE‘S ANOMALY
893
with coarctation of aorta; particular reference to infancy. Circulation 1970;41:1067-75. 7. Shalev Y, Caspi A. Complete atrioventricular block in Shone’s syndrome. Clin Cardiol 1987;10:65-7. 8. Cot0 EO, Judez VM, Juffe A, et al. Supravalvar stenotic mitral ring. A new case with surgical correction. J Thorac Cardiovasc Surg 1976;71:537-9. 9. Konno S, Imai Y, et al. A new method for prosthetic valve replacement in congenital aortic stenosis with hypoplasia of the aortic valve ring. J Thorac Cardiovasc Surg 1975;70: 909-15. 10. Campbell DB, Waldhausen JA. The Konno procedure for enlargement of the aortic root. J Cardiac Surg 1986;1:69-78. 11. Misbach GA, Turley K, Ullyot DJ, Ebert PA. Left ventricular outflow enlargement by the Konno procedure. J Thorac Cardiovasc Surg 1982;84:696-703.
DISCUSSION D R KEVIN TURLEY (San Francisco, CA): I congratulate Bolling and associates for the presentation and the excellent results in this difficult group of patients. As noted, since the first description of the developmental complex of parachute mitral valve, supravalvar ring of the left atrium, subaortic stenosis, and coarctation of the aorta by John Shone and associates in 1963, relatively little has been written about the natural history of the patients who had surgical intervention. However, quite a bit has been contributed to the literature in recent years on the problems of mitral valve abnormalities, especially in the neonate. I think critical to the discussion of this report is an understanding of the patient population and the problems associated with this complex. In the group presented, only 19 patients, 63%, had the full Shone’s syndrome, and in the manuscript it is noted that 7 patients actually had operations on the mitral valve-three initial valvotomies, two second-operation valvotomies, placement of one bioprosthesis, and placement of one mechanical valve. Among those 7 patients, there were four deaths. All four deaths happened at the second operation. These patients, as noted, were representative of the pure Shone’s complex. There were 22 patients with supravalvar mitral rings and 26 patients with mitral valve abnormalities, including parachute mitral valve, fused chordae, and the single papillary muscle. However, only 7 patients had operations on the mitral valve. I think we would like to know more about the other 19 patients with mitral valve abnormalities and the 17 with supravalvar rings and what happened to these valve structures over time.
Further, coarctation was a major problem in this series, as it is for all patients with this syndrome, and 3 of the patients finally received left subclavian-descending aortic grafts. Dr Bolling, could you tell us if there was anything special about the problems of these patients at reoperation for coarctation that necessitated this approach? I agree with the conclusion of Bolling and associates that critical to this problem is the degree of mitral valve disease. I think with both an aggressive surgical approach, as they demonstrated, tailored to the other lesions, and improved techniques of mitral reconstruction, gratifying results can be achieved.
D R BOLLING: The critical point to be noted in patients with Shone’s anomaly is the degree of mitral valve obstruction. Less than half of these patients had mitral valve procedures, and that is one of the problems with a historical review. I think we would agree completely with Dr Turley that our approach to the mitral valve in patients operated on now would be more aggressive. It is interesting to note that in the initial article by Shone and co-workers, which was an autopsy series of 8 patients, 3 of the patients had had previous mitral commissurotomies, and that the authors thought that mitral commissurotomy was not sufficient to repair the mitral disease. In response to the question about left subclavian-descending aortic bypass grafting, these patients are in a historical group, and patients seen more recently would have coarctation repair by other methods. Those patients were, of course, in the group with recurrent coarctation.
Shone‘s anomaly, a congenital cardiac malformation, consists of multiple levels of left heart obstruction including supravalvar mitral ring, parachute mitral valve, subaortic stenosis, and coarctation. The prognosis for patients with Shone’s anomaly is poor. To assess operative results and late outcome, we reviewed the records of 30 consecutive patients seen with Shone’s anomaly at our institution between 1966 and 1989. Anatomical diagnoses in these patients were supravalvar mitral ring (22 patients), mitral valve abnormalities including parachute mitral valve, fused chordae, or single papillary muscle (26 patients), subaortic gradients (26 patients), and coarctation (29 patients). Nineteen patients had all four lesions. Other common defects were bicuspid aortic valve (19 patients) and ventricular septal defect (20). Two patients were treated medically. The other 28 patients required 84 operative procedures with 18 patients undergoing more than one procedure. Operations included coarctation repair (28 patients), mitral valve repair or replacement (111, ventricular septal defect closure (8),
subaortic resection (8), and complex left ventricular outflow tract reconstruction or bypass (4). Age at first operation ranged from 7 days to 7 years (median age, 3 months). There were no operative deaths at the first operation. However, mortality rose to 24% (4117) after the second operation. All operative deaths were secondary to severe mitral valve disease. The survivors have been followed from 1 to 16 years (mean follow-up, 6 f 1 years). There were no late or sudden deaths. Morbidity has included stroke (l), gastrointestinal bleeding (21, permanent heart block ( l ) , and persistent congestive heart failure (6). We conclude that although patients with Shone’s anomaly are seen with a wide spectrum of major anatomical and hemodynamic abnormalities, a satisfactory long-term outcome is possible for most. Operative mortality is adversely affected by the severity of mitral valve disease and the need for multiple operative procedures.
S
Material and Methods
ince the first description of the developmental complex of parachute mitral valve, supravalvar ring of the left atrium, subaortic stenosis, and coarctation of the aorta by Shone and associates [l]in 1963, little clinical experience has accumulated about the operative therapy or late follow-up of patients with Shone’s anomaly. Cardiologists and cardiac surgeons have reported patients with Shone‘s complex to have an unsatisfactory operative outcome and a poor long-term prognosis. Although patients with Shone’s anomaly tend to be grouped together in terms of therapeutic approach, these patients actually demonstrate a wide spectrum of developmental malformations. We postulated that if the operative approach to these patients is tailored and individualized, the long-term outcome might be better. To assess the operative results and the late outcome with such an approach, we retrospectively reviewed the data from 30 consecutive patients with the diagnosis of the developmental complex of Shone’s anomaly.
Presented at the Twenty-fifth Anniversary Meeting of The Society of Thoracic Surgeons, Baltimore, MD, Sep 11-13, 1989. Address reprint requests to Dr Bolling, University of Michigan Hospitals, 1500 E Medical Center Dr, 2120 TC/0344, Ann Arbor, MI 48109.
0 1990 by
The Society of Thoracic Surgeons
(Ann Thorac Surg 1990;49:887-93)
The records of 30 consecutive patients seen at The University of Michigan between 1966 and 1989 were examined for the severity of Shone‘s anomaly, the operative approach, and the eventual late outcome. Reviewed at this time were operative notes, cardiac catheterization reports, inpatient notes, and referring letters. The patients were followed up by a combination of review of outpatient clinic notes and consultation with each patient’s private physician. There were 16 female and 14 male patients. Height and weight at the first operation were 74 2 9 cm and 8.0 t 1.7 kg, respectively. Most patients fell between the fifth and tenth percentile for height and weight at the time of first presentation. Seventeen of the patients were first seen at The University of Michigan in frank congestive heart failure (CHF) necessitating hospitalization. Sixteen of the 17 were on a regimen of digitalis and a diuretic at the time of admission. The condition of 2 additional patients was well controlled with digitalis and a diuretic without evidence of CHF at the time of first admission. Five patients had cardiomegaly on the chest roentgenogram without physical signs of CHF. Four patients had an unremarkable chest roentgenogram and only a murmur on physical 0003-4975/90/$3.50
888
BOLLING ET AL SHONE'S ANOMALY
Ann Thorac Surg 1990;49:887-93
examination, and 1patient with a normal chest roentgenogram was cyanotic on admission. Twelve of the 18 patients who underwent a second operative repair were in CHF. Nine of these 12 patients were receiving digoxin, a diuretic, a vasodilator, or a combination of these before admission. All patients had systolic ejection murmurs greater than 3/6. Four associated syndromes were detected in 3 patients: adrenogenital syndrome, Turner's syndrome, phocomelia, and "floppy baby syndrome." Normal sinus rhythm was demonstrated in 28 patients, and the mean axis was 93 k 8 degrees. One patient was noted to have first-degree heart block. Seven patients had electrocardiographic evidence of left ventricular hypertrophy, 7 patients showed right ventricular hypertrophy, and 9 patients demonstrated biventricular hypertrophy.
Catheterization Data Right and left heart catheterization data obtained in 26 children were compared. Two patients could not undergo left heart catheterization because of technical difficulties, and the records of 2 patients were incomplete. The pressures (mean ? standard error of the mean, systolic/diastolic) for each cardiac chamber were as follows: right atrium, 5 k 1 mm Hg; right ventricle, 62 k 7/7 k 1 mm Hg; pulmonary artery, 60 5/24 & 3 mm Hg; left ventricle, 135 ? 8/13 k 2 mm Hg; ascending aorta, 114 k 7/67 ? 4 mm Hg; and descending aorta, 86 ? 5/60 f 4 mm Hg. Systolic gradients at the area of subaortic obstruction ranged from 10 to 100 mm Hg (mean gradient, 21 mm Hg) and coarctation gradients ranged from 12 to 75 mm Hg (mean gradient, 28 mm Hg). The oxygen saturation was as follows: right atrium, 72% k 1%;right ventricle, 75% ? 1%;pulmonary artery, 76% 5 2%; left ventricle, 94% k 1%;and aorta, 89% k 3%.
*
Anatomy Anatomical evaluation was done in all patients by echocardiography, cardiac catheterization, or both, and then confirmed at operation. The aspects of Shone's anatomy for each patient are shown in Table 1. A diagrammatic representation of the four obstructive lesions in Shone's anomaly is shown in Figure 1. Coarctation of the aorta, found in 29% of the 30 patients, was the most common defect. Subaortic stenosis due to a fibromuscular hypertrophy (25 patients) or a subvalvar membrane (1 patient) was also observed. Abnormalities of the mitral valve were seen in 26 patients and included mitral stenosis, fused chordae, single papillary muscle, and parachute mitral valve apparatus. A supravalvar mitral ring was identified in 22 patients. Nineteen patients had all four components of Shone's anomaly. These were the patients with so-called pure Shone's anomaly. Other associated defects included ventricular septa1 defect (VSD) (20 patients), patent ductus arteriosus (12), and bicuspid aortic valve (19). There were 2 patients with additional left-sided malformations noted at catheterization. One patient had a hypoplastic left ventricle and the other, a severely hypoplastic aortic arch. Other valvar abnormalities included
moderate tricuspid regurgitation (3 patients) and mild pulmonary stenosis (2).
Operative Data Of the 30 patients with Shone's anomaly, 28 underwent operative intervention. The operative procedures are summarized in Table 2. Two patients were treated medically. One patient, an 8-year-old child, was considered to have only minimal hemodynamic abnormalities, despite the presence of all four lesions of Shone's anomaly. The other patient, a 3-month-old infant, was considered to have defects that could not be corrected by operation. Age at the first operation ranged between 7 days and 7 years (mean age, 11 months; median age, 3 months). Nineteen patients had the first operation before 1year of age. There were 42 procedures performed at the first operation in 28 patients, an average of 1.5 2 0.2 per patient. The most common procedure was coarctation repair (23 patients), followed by ligation of a patent ductus arteriosus (10). Other procedures at first operation included mitral valvotomy (3), aortic valvotomy (2), and atrial septostomy, pulmonary artery banding, subaortic resection, and a Blalock-Taussig shunt (1 each). Aortic cross-clamp time was 27 ? 4 minutes for procedures requiring cardiopulmonary bypass. The second operative procedure, necessary in 18 patients, was performed at a mean age of 46 months (range, 1 month to 8 years). The average number of procedures was 1.6 k 0.4 (28 procedures in 17 patients). These operations included VSD repair (7 patients), reoperation for coarctation (4), aortic valvotomy (3), mitral valvotomy (2), subaortic resection of a muscular band (n = 2), and left ventricular outflow tract myotomy, excision of a subaortic membrane, placement of a Hancock bioprosthesis in the mitral position, descending aorta-main pulmonary artery shunt, apicoaortic conduit, pulmonary artery banding, Konno aortoventriculoplasty with St. Jude aortic valve replacement and St. Jude mitral valve replacement, Omniscience mitral valve replacement, resection of a right ventricular muscle band, placement of a permanent pacemaker, and tricuspid valvotomy (1 each). A third operation was performed in 6 patients at a mean age of 52 months (range, 6 months to 7 years). These procedures included right ventricular outflow tract enlargement (2 patients), reoperation for coarctation (2), replacement of an apicoaortic conduit, and placement of a left subclavian artery-descending aortic shunt. Three patients required a fourth operation at 2, 7, and 9 years of age. These procedures included placement of a subclavian artery-descending aortic shunt, pacemaker placement, and reoperation for coarctation. One patient required a fifth operation, a reoperation for recurrent subaortic stenosis, previously resected when the patient was 2 years of age. Six patients underwent reoperation for recurrent coarctation. Two of them required shunting from the left subclavian artery to the descending aorta because of inadequate flow from the left ventricle to the descending aorta owing to restriction at the level of the coarctation. No patient had more than one surgical revision of the coarctation, and 2 patients had balloon angioplasty for
BOLLINGET AL SHONE'S ANOMALY
Ann Thorac Surg 1990;49:887-93
889
Table 1. Summary of Anatomical Data for 30 Patients With Shone's Anomaly Lesions in Shone's Anomaly Patient No.
Left Atrial Ring
Mitral Valve Abnormality
Subaortic Obstruction
Coarctation
All Four
1 2 3 4
No No Yes Yes
Stenosis Normal Fused chordae Stenosis
No Yes Yes No
Yes Yes Yes Yes
No No Yes No
5 6
Yes Yes
Stenosis All to 1 papillary
No Yes
Yes Yes
No Yes
7 8 9 10 11 12 13 14 15 16 17 18
No Yes Yes Yes No No Yes No Yes No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes
Fused chordae Stenosis Stenosis Fused chordae Normal Normal Parachute mitral Stenosis/regurgitation Stenosis Parachute mitral All to 1 papillary Stenosis Stenosis Stenosis Stenosis Fused chordae All to 1 papillary All to 1 papillary Fused chordae All to 1 papillary Parachute mitral Stenosis Normal Stenosis
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes
No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
No Yes Yes Yes No No Yes No Yes No Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes No Yes
19
20 21 22 23 24 25 26 27 28 29 30
ASD = atrial septa1 defect; septa1 defect.
Bi Ao = bicuspid aortic valve;
FDA = patent ductus arteriosus;
this problem. Five patients with coarctation underwent attempts at balloon angioplasty that were considered unsuccessful. Two patients underwent reoperation for recurrent subaortic obstruction that had previously been resected, and 1 patient underwent mitral valve replacement 1 month after a failed mitral commissurotomy. In total, 84 operative procedures were performed in 28 patients with an average of 2.8 separate procedures per patient occurring at an average of 2.4 operations per patient.
Results Mortality There were no operative deaths associated with first operations. All four deaths occurred in the hospital after
Other Anomalies VSD, PDA PDA, PFO, Bi Ao VSD, Bi Ao VSD, PDA, PFO, ASD, hypoplastic arch, Bi Ao None VSD, PDA, PFO, ASD, hypoplastic left ventricle Bi Ao VSD, Bi Ao VSD, Bi Ao VSD, PDA, Bi Ao VSD, Bi Ao VSD, Bi Ao PDA, Bi Ao VSD, PDA, PFO, Bi Ao PDA None VSD, Bi Ao VSD, PDA, PFO, ASD VSD, PDA None PFO, Bi Ao VSD, Bi Ao Bi Ao VSD, PDA, PFO VSD, PDA VSD Bi Ao VSD, Bi Ao VSD, Bi Ao VSD, Bi Ao
PFO = patent foramen ovale;
VSD = ventricular
the second operation (4/17).The first operative death was attributed to low cardiac output after Hancock bioprosthetic valve placement in the mitral position and VSD closure in a child who remained in severe CHF 1 month after a coarctation repair and PDA ligation. This child was 3 months old at the time of death. The second postoperative death occurred in a 3year-old child after coarctation repair and a systemicpulmonary artery shunting procedure, done 2 weeks after an atrial septostomy and mitral valvotomy. The third postoperative death was in a 4-year-old girl who underwent a Konno aortoventriculoplasty with St. Jude prosthetic valve placement in both the aortic and mitral positions. This patient had had a previous flap angioplasty repair of coarctation immediately after birth and remained in severe CHF despite medications. Postopera-
890
BOLLINGET AL SHONE'S ANOMALY
Ann Thorac Surg 1990;49:887-93
Fig 1. The four obstructive anomalies forming Shone's anomaly: supravalvar ring of the left atrium, parachute mitral value, subaortic stenosis, and coarctation of the aorta, in that order according to the direction of blood flow. (Reprinted from [I] by permission of the American lournal of Cardiology.)
tive death was attributed to low cardiac output from right coronary artery occlusion. The fourth patient who died had phocomelia and floppy baby syndrome and was returned to the hospital at age 23 months for a second operation. The second operation included closure of an atrial septa1 defect and unbanding of the pulmonary artery. At that time, the patient had a restrictive VSD with suprasystemic right ventricular pressures and severe desaturation. Aplasia of the bone marrow and severe thrombocytopenia developed and ultimately led to pulmonary hemorrhage and death. Three of 4 patients who died had all four anatomical anomalies of Shone's. There have been no late deaths. The aortic cross-clamp time for the first operation requiring cardiopulmonary bypass was not significantly different between survivors and nonsurvivors (23 & 8 minutes versus 40 5 minutes, respectively). When catheterization data of operative survivors were compared with those of nonsurvivors, the only significant differences were for pulmonary artery systolic and diastolic pressures. The pulmonary artery systolic pressures 14 mm Hg and were 62 & 2 mm Hg versus 102 pulmonary artery diastolic pressures, 22 3 mm Hg versus 39 f 9 mm Hg for survivors and nonsurvivors, respectively (both, p < 0.05).
*
*
*
Morbidity There was one serious postoperative complication resulting in a prolonged hospital course after the first operation. It was a sternal wound infection, which resolved with
intravenous antibiotics and secondary closure. No organisms were ever identified. In addition, 1 patient was transferred to our institution with the symptoms of peritonitis and guaiac-positive stools. This patient had all four abnormalities of Shone's and was in florid CHF at the time of presentation. Necrotizing enterocolitis developed after the cardiac operation, and the patient required prolonged hyperalimentation and hospitalization. No abdominal procedure was necessary as a result of this complication. The incidence of major morbidity increased with each succeeding operation. After the second operation, 1 patient sustained a series of postoperative complications, which included a posterior myocardial infarction, reexploration for bleeding, an upper gastrointestinal bleed, stroke, and deficiency of coagulation factors I1 and V. This patient had replacement of the apicoaortic conduit and enlargement of the right ventricular outflow tract and is now doing well 19 months after operation. A series of conduction abnormalities due to complete heart block were noted in 1 patient after operation and resulted in placement of a permanent pacemaker. One patient each had right bundle-branch block, right ventricular conduction delay, and left bundle-branch block, but these conditions did not necessitate placement of a permanent pacemaker.
Follow-up Data Of the 30 patients identified with Shone's anomaly, 4 patients died and no follow-up was available for 1patient. Of the 2 patients treated medically, 1 returned to Japan after the diagnosis of Shone's anomaly was made, and she was lost to follow-up. This young Oriental girl had adrenogenital syndrome as well as Shone's anomaly. The other medically treated patient had Turner's syndrome as well as mild Shone's anomaly, and is now working as a secretary. Of the 28 patients who underwent surgical intervention, 24 were eligible for follow-up after their last operative procedure. The average length of follow-up is 5.7 & 1.1 years after the most recent operation. The latest New York Heart Association activity level was class I for 82%, class I1 for 12%, and class I11 for 6%. Seven patients have persistent cardiomegaly on chest roentgenograms. Six patients still require medical therapy for control of persistent episodes of CHF. At the most recent postoperative catheterization, the mean follow-up gradient through the area of repaired aortic coarctation was 16 2 4 mm Hg (7 patients) and across the aortic outflow tract, 33 10 mm Hg (5 patients). Two patients still have apicoaortic conduits in place. Only 4 patients were reported to have a residual gradient across the mitral valve (mean gradient, 5 1 mm Hg). Other reported valvar abnormalities at follow-up were mild mitral regurgitation in 3 patients and moderate tricuspid regurgitation in 1 patient. Although these patients with Shone's anomaly remain quite small physically, success in school was at least average in this group based on follow-up reports. Furthermore, only 3 of the patients in the follow-up group were
*
*
BOLLING ET AL SHONE‘S ANOMALY
Ann Thorac Surg 1990;49:887-93
891
Table 2. Summary of Operative Procedures From Most Recent to lnitial Patient No. 1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
30
Operative Repairs and Procedures
1: Redo CoA repair with Dacron; 2: VSD closure; 3: CoA repair and PDA ligation 1: CoA repair with Gore-Tex patch and PDA ligation 1: Mitral valvotomy; 2: CoA balloon angioplasty for restenosis; 3: CoA repair 1: Redo CoA repair with Dacron; 2: CoA repair 1: CoA repair with LSCA patch No repair 1: Open aortic and mitral commissurotomies 1: CoA repair (end to end) 1: Subaortic resection of muscle band, VSD closure; 2: CoA repair 1: Aortic valvotomy, LV outflow tract myomectomy, VSD closure; 2: subaortic membrane excision; 3: FDA ligation, CoA repair 1: Resection of subaortic restenosis; 2: redo CoA repair; 3: resection and patch of RV outflow tract obstruction; 4: VSD closure; 5: PDA ligation, CoA repair 1: CoA repair and PDA ligation 1: CoA repair and PDA ligation 1: Hancock bioprosthetic replacement of mitral valve, VSD closure; 2: CoA repair, PDA ligation 1: CoA repair, Potts shunt; 2: atrial septostomy, mitral valvotomy No repair 1: CoA repair with LSCA 1:PA banding, CoA repair; 2: right Blalock-Taussig shunt, atrial septostomy 1: VSD closure; 2: CoA repair with LSCA, PDA ligation 1: RV outflow tract enlargement, replacement of apicoaortic conduit; 2: LV apex-aorta conduit; 3: CoA repair 1: Mitral valvotomy; 2: CoA repair 1: Redo CoA repair with Dacron; 2: CoA repair with LSCA 1: CoA repair with LSCA; 2: aortic commissurotomy, resection of subaortic stenosis 1: CoA repair with Gore-Tex, PDA ligation 1: Konno aortoventriculoplasty with St. Jude replacement of aortic and mitral valves, CoA balloon angioplasty for restenosis; 2: CoA repair with LSCA 1: PA banding, CoA repair 1: CoA repair (end to end) 1: LSCA-descending aortic shunt; 2: CoA repair (end to end); 3: MVR with No. 17 Omniscience; 4: mitral commissurotomy 1: Pacemaker placement for third-degree heart block; 2: resection of subaortic band, resection of RV band, patch closure of VSD; 3: Ao valvotomy, tricuspid valvotomy, PDA ligation; 4: CoA repair with LSCA 1: PDA, CoA repair with LSCA; 2: Closure of VSD
CoA = coarctation of aorta; LV LSCA = left subclavian artery; artery; FDA = patent ductus arteriosus; RV = right ventricular;
=
considered to be ”minimally” handicapped or to have learning disabilities.
Comment More than 25 years have passed since Shone and colleagues [ 11 first described the developmental complex of parachute mitral valve, supravalvar ring of the left atrium, subaortic stenosis, and coarctation of the aorta. In that original article (8 patients), only 2 patients exhibited all four aspects of Shone’s anomaly and could therefore be considered to have pure cases of Shone’s anomaly. In our series, 19 patients were seen with all four aspects of the anomaly and therefore were considered to have pure cases. The other 11 can be regarded as having forme fruste
MVR = mitral valve replacement; left ventricular; VSD = ventricular septa1 defect.
PA = pulmonary
of Shone’s anomaly, most missing a supravalvar mitral ring. In the description of Shone and associates [l],it was noted that mitral valve obstruction appeared to be the most critical problem associated with the anomaly, and mitral commissurotomy alone (performed in 3 of the patients) was not adequate to relieve the mitral valve obstruction. Shone and colleagues quite farsightedly made a hypothetical case for excision of the supravalvar ring and replacement of the mitral valve with a prosthesis as the best method of overcoming the major mitral valvar and subvalvar obstructions. Our present study confirms the belief that severity of the mitral valve obstruction correlates inversely with long-term outcome and that operative mortality in pa-
892
BOLLING ET AL SHONE'S ANOMALY
tients with Shone's anomaly is adversely affected by the degree of mitral valve disease. The patients who demonstrated substantially elevated pulmonary artery pressures were those with the worst mitral obstruction and had, in general, the poorest outcome. These patients also appeared to have severe and progressive left ventricular outflow tract obstruction. Kieffer and associates [2] noted an association between reversed ventricular septal curvature seen on two-dimensional echocardiography with primary pulmonary hypertension and Shone's anomaly. Indeed, our patients with severe mitral obstructive disease as a result of Shone's anomaly did have secondary pulmonary hypertension, which conceivably could have led to increased reversed ventricular septal curvature and exacerbation of subaortic obstruction. Rosenquist [3] described the results of pathological study of 53 heart specimens in which the primary diagnosis was coarctation of the aorta with normally related great vessels. Rosenquist noted a surprisingly high incidence of obstructive mitral valve involvement in these specimens. Furthermore, the mitral valve abnormalities usually involved the subvalvar apparatus and demonstrated a wide spectrum of anomalies from hypoplastic but otherwise normal mitral valves to severe parachute mitral stenosis. In addition, that report [3] noted three abnormalities in the mitral valves of patients with Shone's anomaly, findings subsequently confirmed by others [4]. These abnormalities involved the mitral subvalvar apparatus and included fusion of the papillary muscles, fusion of a papillary muscle to the ventricular wall, and true parachute mitral valve with only one papillary muscle. Varying degrees of all these forms of subvalvar mitral abnormality were noted in our patients (see Table 1). In a clinicopathological conference report demonstrating the case of a patient with Shone's anomaly [ 5 ] , physicians were cautioned to exercise an awareness of the possibility of other left-sided obstructive lesions that could be present in the symptomatic infant with coarctation. The correlation of obstructive left-sided anomalies associated with coarctation of the aorta was confirmed by Becker and co-workers [6] 2 years later. They believed that Shone's anomaly should be excluded in those patients who continue to have signs of CHF after operative repair of the coarctation. This correlates very well with our present group of patients, many of whom demonstrated continued CHF after the first operation, which was most commonly coarctation repair and ligation of patent ductus arteriosus. As recently as 1987 in a report of spontaneous complete atrioventricular conduction block in Shone's anomaly [71, it was stated that the diagnosis of Shone's anomaly is usually made during infancy or early childhood. However, the authors then noted that most patients with Shone's anomaly are dead soon after the diagnosis is made or soon after an attempt at surgical repair. Many cardiac surgeons and cardiologists agree with this dismal outlook, and therefore, the prevailing sentiment is that the prognosis for these patients is poor. However,
Ann Thorac Surg 1990;49:887-93
few data are available to document the outcome of patients with Shone's anomaly. Although there have been isolated case reports [8] describing the operative approach to the different lesions that make up Shone's anomaly, no overall operative strategy has been developed. Our operative approach to these patients includes early repair of coarctation and ligation of patent ductus arteriosus. Repair of other lesions including VSD can generally be postponed. The majority of VSDs in our series were restrictive and closed spontaneously after coarctation repair. The surgical options for the left ventricular outflow tract obstruction must be tailored to the individual patient because of the wide variability in anatomy. When the aortic annulus is adequate and the obstruction is limited to the subaortic area, septoplasty is an effective procedure. If the annulus is hypoplastic, we favor aortoventriculoplasty (Konno procedure) [9, 101 over left ventricular apex-descending aortic shunts. Aortoventriculoplasty has been used for tunnel subaortic obstruction, and the incision in the interventricular septum can easily be taken well into the left ventricular cavity, effectively relieving the obstruction. The Konno procedure allows the aortic valve to remain in the normal anatomical position with normal flow characteristics and permits the placement of adequately large valves, despite the overall small body size of the patient. This procedure has many other advantages compared with apicoaortic conduit procedures, including decreased arrhythmias and preservation of ventricular function [ll]. The operative treatment of the mitral lesions noted in these patients remains difficult because of the highly variable nature of these malformations and the differing degrees of subvalvar involvement. However, valve repair should be considered whenever feasible in these young patients before mitral valve replacement is performed. The frequency of multiple operations in patients with Shone's anomaly (18 of 30 patients undergoing more than one procedure) makes this group a formidable technical challenge. Furthermore, the high risk of subsequent operations, particularly in those patients with mitral valve involvement, can discourage even the most aggressive surgeon. Finally, although our patients have had a satisfactory long-term outcome, 4 of 24 children have a prosthetic valve in place and must undergo further reoperative procedures to replace bioprosthetic valves as they degenerate or suffer the consequences of anticoagulation, which have led to substantial gastrointestinal bleeding in 2 patients. In summary, our report demonstrates that patients with Shone's anomaly represent a wide spectrum of major anatomical and hemodynamic abnormalities. An aggressive surgical approach tailored to each patient's individual anatomy can result in a satisfactory outcome for most of these patients.
References 1. Shone JD, Sellers RD, Anderson RC, Adams P, Lillehei CW, Edwards JE. The developmental complex of "parachute mi-
Ann Thorac Surg 19’30;49887-93
tral valve,” supravalvular ring of left atrium, subaortic stenosis, and coarctation of aorta. Am J Cardiol 1963;11:71&25. 2. Kieffer RW, Hutchins GM, Moore GW, Buckley BH. Reversed septa1 curvature. Association with primary pulmonary hypertension and Shone‘s syndrome. Am J Med 1978; 66:831-5. 3. Rosenquist GC. Congenital mitral valve disease associated with coarctation of the aorta. A spectrum that includes parachute deformity of the mitral valve. Circulation 1974; 49:98593. 4. Tandon R, Moller JH, Edwards JE. Anomalies associated with the parachute mitral valve: a pathologic analysis of 52 cases. Can J Cardiol 1986;2:27%81. 5. Noren GR. Coarctation of aorta as part of Shone syndrome. Minn Med 1968;9:1617-27. 6. Becker AE, Becker MJ, Edwards JE. Anomalies associated
BOLLING ET AL SHONE‘S ANOMALY
893
with coarctation of aorta; particular reference to infancy. Circulation 1970;41:1067-75. 7. Shalev Y, Caspi A. Complete atrioventricular block in Shone’s syndrome. Clin Cardiol 1987;10:65-7. 8. Cot0 EO, Judez VM, Juffe A, et al. Supravalvar stenotic mitral ring. A new case with surgical correction. J Thorac Cardiovasc Surg 1976;71:537-9. 9. Konno S, Imai Y, et al. A new method for prosthetic valve replacement in congenital aortic stenosis with hypoplasia of the aortic valve ring. J Thorac Cardiovasc Surg 1975;70: 909-15. 10. Campbell DB, Waldhausen JA. The Konno procedure for enlargement of the aortic root. J Cardiac Surg 1986;1:69-78. 11. Misbach GA, Turley K, Ullyot DJ, Ebert PA. Left ventricular outflow enlargement by the Konno procedure. J Thorac Cardiovasc Surg 1982;84:696-703.
DISCUSSION D R KEVIN TURLEY (San Francisco, CA): I congratulate Bolling and associates for the presentation and the excellent results in this difficult group of patients. As noted, since the first description of the developmental complex of parachute mitral valve, supravalvar ring of the left atrium, subaortic stenosis, and coarctation of the aorta by John Shone and associates in 1963, relatively little has been written about the natural history of the patients who had surgical intervention. However, quite a bit has been contributed to the literature in recent years on the problems of mitral valve abnormalities, especially in the neonate. I think critical to the discussion of this report is an understanding of the patient population and the problems associated with this complex. In the group presented, only 19 patients, 63%, had the full Shone’s syndrome, and in the manuscript it is noted that 7 patients actually had operations on the mitral valve-three initial valvotomies, two second-operation valvotomies, placement of one bioprosthesis, and placement of one mechanical valve. Among those 7 patients, there were four deaths. All four deaths happened at the second operation. These patients, as noted, were representative of the pure Shone’s complex. There were 22 patients with supravalvar mitral rings and 26 patients with mitral valve abnormalities, including parachute mitral valve, fused chordae, and the single papillary muscle. However, only 7 patients had operations on the mitral valve. I think we would like to know more about the other 19 patients with mitral valve abnormalities and the 17 with supravalvar rings and what happened to these valve structures over time.
Further, coarctation was a major problem in this series, as it is for all patients with this syndrome, and 3 of the patients finally received left subclavian-descending aortic grafts. Dr Bolling, could you tell us if there was anything special about the problems of these patients at reoperation for coarctation that necessitated this approach? I agree with the conclusion of Bolling and associates that critical to this problem is the degree of mitral valve disease. I think with both an aggressive surgical approach, as they demonstrated, tailored to the other lesions, and improved techniques of mitral reconstruction, gratifying results can be achieved.
D R BOLLING: The critical point to be noted in patients with Shone’s anomaly is the degree of mitral valve obstruction. Less than half of these patients had mitral valve procedures, and that is one of the problems with a historical review. I think we would agree completely with Dr Turley that our approach to the mitral valve in patients operated on now would be more aggressive. It is interesting to note that in the initial article by Shone and co-workers, which was an autopsy series of 8 patients, 3 of the patients had had previous mitral commissurotomies, and that the authors thought that mitral commissurotomy was not sufficient to repair the mitral disease. In response to the question about left subclavian-descending aortic bypass grafting, these patients are in a historical group, and patients seen more recently would have coarctation repair by other methods. Those patients were, of course, in the group with recurrent coarctation.
