- Email: [email protected]
Discrete Membranous Subaortic Stenosis: Improved Results After Resection and Myectomy
Discrete Membranous Subaortic Stenosis: Improved Results After Resection and Myectomy Samuel T. Rayburn, MD, Donald E. Netherland, MD, and Bobby J. Heath, MD Division of Cardiothoracic Surgery, University of Mississippi Medical Center, Jackson, Mississippi
Background. Despite an adequate resection, a significant recurrence rate is encountered in patients undergoing operation for discrete membranous subaortic stenosis. The fibrous membrane and hypertrophied myocardium commonly are removed, but because of the involved risks, the resection may be inadequate and contribute to the recurrence rate. Methods. A review of the cases of 23 patients undergoing operation for discrete membranous subaortic stenosis from 1980 to 1994 was undertaken. Fourteen patients (61%) had coexisting cardiac lesions, all of which were concomitantly repaired. Results. The left ventricle–aorta gradient decreased from a preoperative mean of 63.39 6 7.63 mm Hg to 15.17 6 3.06 mm Hg postoperatively (p < .001) during a
mean follow-up of 3.32 6 0.58 years. Aortic insufficiency decreased postoperatively in 8 patients (34.8%), remained unchanged in 6 patients (26.1%), and showed only insignificant progression in 4 patients (17.4%). There were no early deaths, and the single late death was not cardiac related. No patient had development of endocarditis or heart block or required a pacemaker. One patient (4.3%) had a recurrence, which required reoperation. Conclusions. Our results suggest that aggressive myectomy in concert with membrane resection constitutes safe treatment for discrete membranous subaortic stenosis and is associated with low rates of endocarditis, recurrence, and progression of aortic insufficiency. (Ann Thorac Surg 1997;64:105–9) © 1997 by The Society of Thoracic Surgeons
C
the time of surgical correction for relief of LVOT stenosis, both to prevent the associated sequelae of DMSS and to lessen the likelihood of recurrence. Our results with this technique over a 15-year period are the basis of this report.
ongenital aortic stenosis results from an obstruction of the left ventricular outflow tract (LVOT). The lesion is classified as valvular, subvalvular, or supravalvular, and it can be associated with a variety of other cardiac anomalies. Subvalvular aortic stenosis accounts for 8% to 30% of patients with congenital LVOT obstruction and generally occurs in one of two forms: it occasionally is secondary to a circumferential fibromuscular tunnel, but it more commonly results from a discrete membrane immediately below the aortic valve [1]. The discrete form of subvalvular aortic stenosis (DMSS) has been classified as either a thin, fibrous membrane (type I) or a thicker fibromuscular band (type II) [2]. In our experience, however, all clinically significant lesions have some degree of obstructing hypertrophic myocardium of the LVOT. The natural history of DMSS is one of progressive obstruction, and it is associated with the sequelae of aortic insufficiency (AI), endocarditis, and recurrent obstruction after resection. A major factor in recurrent LVOT obstruction is believed to be inadequate relief of the obstruction at operation. Although the fibrous membrane is universally excised, some groups [3–5] advocate concomitant selective myectomy to achieve full relief of the LVOT stenosis, whereas others [6] have reported that myectomy adds little to the procedure. We believe that aggressive myectomy is necessary at Accepted for publication Jan 14, 1997. Address reprint requests to Dr Heath, Division of Cardiothoracic Surgery, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216.
© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Material and Methods Data Collection All patients undergoing surgical treatment of DMSS from 1980 to 1994 were included in this study; all patients seen with tunnel subvalvular aortic stenosis (n 5 4) were excluded. The case histories of all patients, including patient and primary physician interviews and medical records, were examined retrospectively. All patients were evaluated preoperatively and followed for variable periods postoperatively at The University of Mississippi Medical Center. No patient was lost to follow-up. Information compiled included the patient’s age, sex, previous cardiac operation, and coexisting cardiac anomalies. In addition to the recording of physical signs and symptoms, electrocardiography, chest roentgenography, echocardiography, and often cardiac catheterization were performed preoperatively and at regular postoperative intervals. All complications, reoperations, and deaths were noted. Statistically, left ventricle–aorta gradient measurements collected from cardiac catheterization and echocardiography were used interchangeably and have been shown in prior studies to correlate well [7].
Surgical Procedure All operations were performed by the same surgeon (B.J.H.) at The University of Mississippi Medical Center under stan0003-4975/97/$17.00 PII S0003-4975(97)00371-8
106
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
dard cardiopulmonary bypass, using cold potassium cardioplegia for myocardial protection. Indications for surgical treatment varied but included left ventricle–aorta gradient greater than 30 to 40 mm Hg, evidence of new or progressive AI, and coexisting cardiac lesions requiring operation. A transverse aortotomy was used; the incision was often carried obliquely into the noncoronary sinus. Appropriate traction on the aortic leaflets provided excellent exposure of the subaortic region. The obstructing membrane was circumferentially excised, and a deep, generous trough of hypertrophied muscle in the ventricular septum between the right and left coronary aortic cusps was resected. Great care was exercised to avoid injury to the conduction tissue between the right and noncoronary cusps and the anterior leaflet of the mitral valve, to which the membrane occasionally was adherent.
Statistical Analysis Results are expressed as a range with the mean 6 the standard error of the mean. Group means were compared with Student’s t test for paired events, and event frequencies with a p value of less than 0.05 were considered significant.
Results Preoperative Data From April 1980 to September 1994, 23 patients with DMSS underwent surgical correction. The group comprised 12 female and 11 male patients ranging in age from 7 months to 51 years (mean age, 13.3 6 2.46 years). Five patients (21.7%) had had a previous cardiac surgical procedure: closure of a patent ductus arteriosus in 1; a closed valvulotomy for critical aortic stenosis as a newborn in 1; repair of aortic coarctation in 1; pulmonary artery banding for type I double-outlet right ventricle, closure of a patent ductus arteriosus, and aortic coarctation repair in 1; and repair of an atrioventricular canal defect with mitral valve replacement in 1. Seven patients (30%) were noted to be symptomatic: 3 had dyspnea on exertion (2 with evidence of mild congestive heart failure); 1 had angina pectoris; 2 experienced both dyspnea on exertion and angina; and 1 patient, previously asymptomatic, was seen in ventricular tachycardia after a syncopal episode. The remaining 16 patients (70%) were asymptomatic, and the diagnosis was made on discovery of their murmurs. No patient had endocarditis preoperatively. On physical examination, all patients were noted to have systolic ejection murmurs, 11 with palpable thrills, and 8 had diastolic murmurs of AI. Preoperative evaluation with echocardiography or cardiac catheterization disclosed left ventricle–aorta gradients ranging from 13 mm Hg to 155 mm Hg (mean gradient, 63.39 6 7.63 mm Hg). Fourteen patients (61%) had documented AI preoperatively (11 in 10 patients, 21 in 2 patients, and 31 in 2 adult patients). Of the 23 patients, 9 (39%) had isolated DMSS, and 14 (61%) had coexisting cardiac lesions, all of which were corrected concomitantly at the time of operation for DMSS. One patient undergoing resection of isolated DMSS had had a previous cardiac operation to repair an
Ann Thorac Surg 1997;64:105–9
atrioventricular canal defect with mitral valve replacement and correction of aortic coarctation. Associated lesions and operations included the following: aortic stenosis requiring commissurotomy in 5 patients, 3 of whom had bicuspid aortic valves; aortic valve replacement in 2, 1 of whom had a dysplastic aortic valve and 1, severe aortic regurgitation that was also associated with a patent ductus arteriosus; closure of a ventricular septal defect in 3 patients, one of which was associated with a patent foramen ovale and one, with right ventricular outflow tract obstruction; right ventricular outflow tract obstruction in 1 patient; right ventricular outflow tract obstruction and a patent ductus arteriosus in 1 patient; patent foramen ovale in 1 patient; and type I doubleoutlet right ventricle in 1 patient. Table 1 summarizes the preoperative and postoperative patient data.
Postoperative Findings Postoperative follow-up was possible in all patients and ranged from 7 months to more than 9 years (mean follow-up, 3.32 6 0.58 years). Nonfatal postoperative complications in 5 patients (21.7%) proved transient. These complications included nonspecific perioperative arrhythmias in 3 and high-output renal failure in 2 adults. There were no early postoperative deaths, and to date, there has been only one late death (4.3%), which followed an episode of acute pancreatitis. The postoperative left ventricle–aorta gradient was significantly decreased at late follow-up (p , 0.001) and ranged from 0 mm Hg to 40 mm Hg (mean gradient, 15.17 6 3.06 mm Hg) (Fig 1). Aortic insufficiency progressed in 4 patients (17.4%) (0 to 11), decreased in 8 patients (34.8%) (11 to 0 in 6 patients, 31 to 21 in 1 patient, and 31 to 0 in 1 patient), and remained unchanged in 6 patients (26%); 5 patients had no evidence of AI throughout the study period. Symptoms were completely relieved in all symptomatic patients (71.4%). After resection, 1 patient had intermittent dyspnea on exertion, and another patient, who was seen initially with ventricular tachycardia, had intermittent arrhythmias. Both were treated medically and are currently asymptomatic. Of the 16 asymptomatic patients, 15 remained without symptoms throughout follow-up. One became symptomatic more than 3 years postoperatively, and her symptoms were relieved after reoperation. Two patients required reoperation: the patient just mentioned, who had a morphologically identical recurrence of DMSS, and 1 patient who required closure of a recurrent ventricular septal defect, which was previously repaired at the time of DMSS resection. No patient had progression to the more severe tunnel form of subaortic stenosis, had development of heart block, required a pacemaker, or had development of endocarditis during the study period.
Comment In 1842, Cheevers [8] recorded the first description of subvalvular aortic stenosis. The pathogenesis of this lesion has proved elusive, but a number of theories have
Ann Thorac Surg 1997;64:105–9
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
107
Table 1. Summary of Patient Data Preoperative Data
Age
LV–Ao Gradient (mm Hg)
1
25 y 8 mo
90
2 3 4 5 6 7 8 9 10 11 12 13
51 y 9 y 3 mo 23 y 11 y 32 y 5y 26 y 16 y 13 y 18 y 5 y 6 mo 2 y 8 mo
40 155 70 130 80 70 50 35 60 110 20 80
14 15 16 17
6 y 10 mo 8y 10 y 3 y 8 mo
30 50 40 13
18 19 20 21 22 23
7 mo 4 y 5 mo 1 y 11 mo 5y 14 y 14 y
90 30 90 20 35 60
Patient No.
AI
Associated Procedures
21 PDA ligation, resection RVOT stenosis 0 None 0 None 31 PDA ligation, AVR 0 None 11 Aortic valvotomy 0 PFO closure 11 Aortic valvotomy 11 None 11 None 11 AVR 0 Resection RVOT stenosis 0 DORV repair, PA band removal 0 None 11 VSD closure 11 None 11 VSD closure, resection RVOT stenosis 31 Aortic commissurotomy 21 Aortic commissurotomy 0 Aortic valvotomy 11 PFO and VSD closure 0 None 11 None
Postoperative Data LV–Ao Gradient at Late Follow-up Follow-up (mm Hg)
AI
Endocarditis
2 y 2 mo
30
21
No
No
1 y 6 mo 8y 9 y 9 mo 8 y 1 mo 4y 8 y 10 mo 3 y 1 mo 2 y 5 mo 2 y 10 mo 3 y 6 mo 4 y 10 m 4 y 6 mo
15 26 16 35 25 24 0 0 0 40 0 0
0 0 0 11 11 11 11 0 0 0 11 11
No No No No No No No No No No No No
No No No No No No No No No No No Yes (recurrent DMSS)
1 y 6 mo 1 y 5 mo 7 mo 1 y 5 mo
0 21 0 0
0 0 11 0
No No No No
No No No
1 y 10 mo 1 y 4 mo 1 y 2 mo 1 y 8 mo 1 y 3 mo 8 mo
30 26 25 0 0 36
21 21 0 0 0 11
No No No No No No
No No No Yes (recurrent VSD) No No
Reoperation
AI 5 aortic insufficiency; AVR 5 aortic valve replacement; DMSS 5 discrete membranous subaortic stenosis; DORV 5 double-outlet right ventricle; LV–Ao 5 left ventricle–aorta; PA 5 pulmonary artery; PDA 5 patent ductus arteriosus; PFO 5 patent foramen ovale; RVOT 5 right ventricular outflow tract; VSD 5 ventricular septal defect.
been proposed to explain its occurrence. The explanations attribute DMSS to congenital [9], inflammatory [8], genetic [10], and acquired causes [11]. On the basis of
Fig 1. Change in left ventricle–aorta (LV-AO) peak gradient from before to after operation.
observations of postoperative recurrence identical to the primary lesion, it has been proposed that the nidus is myocardial and that membrane development is a secondary event [12]. We believe the primary lesion is myopathic, which would account for the cases of DMSS in patients with concomitant right ventricular outflow tract stenosis. The progressive nature of DMSS is readily explained by the Rodbard phenomenon [13]. Rodbard reproduced abnormal flow hemodynamics in an animal model, and in accordance with Bernoulli’s principle of energy conservation, progressively stenotic lesions were produced. This would also appear to account for the lesions associated with other defects that produce abnormal flow patterns, such as atrioventricular canal, ventricular septal defect, and double-outlet right ventricle. No one explanation, however, appears to account for all cases of DMSS, and it seems likely to have multiple potential sources. Discrete subaortic stenosis represents a fixed obstruction in which the LVOT stenosis is caused by a specific anatomic lesion and remains unchanged with b agonists. This is in contrast to the dynamic obstruction of idio-
108
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
Ann Thorac Surg 1997;64:105–9
Table 2. Outcome Data From Reported Series
Reference Katz et al (1977) [14] Wright et al (1983) [5] Cain et al (1984) [12] Hardesty et al (1984) [3] Ashraf et al (1985) [6] Brown et al (1985) [17] Douville et al (1990) [15] Jacobs et al (1994) [16] Rayburn et al (1997) [this report]
Early and Late Deaths (%)
AI (% preop)
Recurrence Requiring Reoperation (%)
Endocarditis (%)
7.7 17 2.2 6 8.2 9.4 3 5.6 4.3
29 55 ? ? 18.4 79 59 ? 61
7.7 14 30 14 15 5.7 9 16.7 4.3
13 12 ? 2.9 2 9 3 ? 0
AI 5 aortic insufficiency.
pathic hypertrophic subaortic stenosis, in which the obstruction is secondary to generalized myocardial hypertrophy of the LVOT and worsens with b agonists. Although these have generally been regarded as separate entities, there appears to be a major interrelationship between the two myopathies. A dynamic component is postulated to be a factor in the residual LVOT stenosis of DMSS after adequate resection and may predispose to recurrent obstruction or progression to the more severe tunnellike stenosis [11, 14]. Although knowledge of the pathogenesis and pathophysiology of the lesion is limited, several aspects of the disease are more clearly defined. There is an established association of DMSS with AI and endocarditis, and there is a substantial recurrence rate after surgical excision. Many of the studies in the recent literature include patients who had operations prior to the routine use of cardioplegia for myocardial protection and echocardiography for follow-up, which allow a more complete anatomic resection and noninvasive detection of postoperative sequelae, respectively. Thus, the usefulness of the data in these studies and their conclusions is limited. Our study confirms several aspects of the natural history of DMSS. All patients with marked DMSS have a prominent systolic murmur, often with a palpable thrill and occasionally in concert with a diastolic murmur of AI. The majority of our patients were asymptomatic (70%) but this varies from study to study [14 –17]. We affirm that DMSS often coexists with a variety of congenital cardiac abnormalities; only 39% of our patients had isolated DMSS. Several complications can result from the surgical treatment of DMSS. Overzealous resection in the ventricular septum can lead to heart block from damage to the conduction system, iatrogenic ventricular septal defect, or damage to the anterior leaflet of the mitral valve. Our data confirm that operative treatment of DMSS with extensive myectomy is safe and effective; we had no permanent complications (including development of heart block or requirement of a pacemaker), no early postoperative deaths, and only one late death. That single death occurred more than 6 years postoperatively and was due to complications of acute pancreatitis in an adult patient with multiple medical problems.
The majority of patients with DMSS have abnormalities of the aortic valve, some of which are congenital, but the majority of which are acquired [11]. Aortic insufficiency is the most common acquired lesion of the aortic valve in DMSS and is believed to result from trauma to the valve leaflets by the abnormal jet flow pattern caused by the lesion [5]. The AI can progress postoperatively despite relief of the LVOT stenosis of DMSS [6], but it has been noted that worsening of the AI in DMSS can be slowed or stopped with adequate operative resection of DMSS [5]. Our data support these conclusions. Aortic insufficiency did increase in 17.4% of our patients, but in all of them, the increase was clinically insignificant (0 to 11). In 34.8% of patients with AI preoperatively, the severity decreased. Two of these 8 patients, however, required aortic valve replacement, 1 for a congenitally dysplastic valve and 1 for 31 acquired AI. The aortic valve always remains a potential site for development of endocarditis in patients with DMSS as a result of valve thickening from the jet flow pattern [11], with a reported incidence of up to 13% [14]. The predisposing factors are high LVOT gradient and AI. As with AI, successful relief of LVOT obstruction is believed to reduce the risk of endocarditis in DMSS [5]. None of our patients had development of endocarditis preoperatively or postoperatively during the course of this study. Recurrent LVOT obstruction after resection is well documented, and reoperation is required at a rate of 6% to 30% in similar series [5, 12, 15–17]. This high rate is credited mainly to inadequate resection at the first operation, but a component of dynamic obstruction is also thought to be responsible [11]. The classification of DMSS lesions by Kelly and associates [2] (type I, thin membrane alone, and type II, membrane with associated hypertrophied myocardium) is somewhat misleading in that the lesion of DMSS is progressive and represents but a point in a spectrum in which tunnel stenosis represents the end point. No patient in our series had a membrane without a substantial degree of myocardial hypertrophy obstructing the LVOT (type II), and although reports of type I DMSS are documented in the literature, we believe they are rarely the cause of major LVOT stenosis. Therefore, both myectomy and membrane resection must be undertaken to achieve full relief of the stenosis. To our
Ann Thorac Surg 1997;64:105–9
knowledge, this concept has never been tested in a prospective, randomized trial, and all conclusions are based on retrospective studies; this is likely the result of the relatively infrequent occurrence of this disease. In the series of Ashraf and colleagues [6] and Cain and coworkers [12], use of myectomy did not significantly decrease the number of recurrences that required reoperation. However, both series report reoperation rates in excess of 15%. Our rate of reoperation for recurrence was lower (4.3%) than those previously reported, and no patient had progression to the more severe form of tunnel stenosis. We think that this and our low incidences of AI progression and endocarditis is a result of our commitment to perform an extensive myectomy to fully relieve the LVOT obstruction (Table 2). Resection of the obstructing membrane in concert with generous myectomy of the LVOT is a safe and effective treatment of DMSS. Our results suggest a significant decrease in postoperative left ventricle–aorta gradient, a low rate of recurrent postoperative LVOT obstruction, insignificant progression of AI, and nonexistent rates of endocarditis development and progression to the more severe form of tunnel subaortic stenosis.
References 1. Kirklin JW, Barratt-Boyes BG. Cardiac surgery. 2nd ed. New York: Churchill Livingstone, 1993:1212–24. 2. Kelly DT, Wulfsberg E, Rowe RD. Discrete subaortic stenosis. Circulation 1972;46:309–22. 3. Hardesty RL, Griffith BP, Mathews RA, et al. Discrete subvalvular aortic stenosis. An evaluation of operative therapy. J Thorac Cardiovasc Surg 1984;118:79– 83. 4. McKay R, Ross DN. Technique for the relief of discrete subaortic stenosis. J Thorac Cardiovasc Surg 1982;84:917–20.
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
109
5. Wright GB, Keane JF, Nadas AS, Bernhard WF, Castan˜eda AR. Fixed subaortic stenosis in the young: medical and surgical course in 83 patients. Am J Cardiol 1983;52:830–5. 6. Ashraf H, Cotroneo MD, Dhar N, et al. Long-term results after excision of fixed subaortic stenosis. J Thorac Cardiovasc Surg 1985;90:864–71. 7. Lima CO, Sahn DJ, Valdes-Cruz LM, et al. Prediction of the severity of left ventricular outflow tract obstruction by quantitative two-dimensional echocardiographic Doppler studies. Circulation 1983;68:348–54. 8. Cheevers N. Observations on the diseases of the orifice and valves of the aorta. Guys Hosp Rep 1842:387– 442. 9. Keith A. Schorstein Lecture on the fate of the bulbis cordis in the human heart. Lancet 1924;2:1267–72. 10. Pyle RL, Patterson DF, Chacko S. The genetics and pathology of discrete subaortic stenosis in the Newfoundland dog. Am Heart J 1976;3:324–34. 11. Somerville J, Stone S, Ross D. Fate of patients with fixed subaortic stenosis after surgical removal. Br Heart J 1980;43: 629– 47. 12. Cain T, Campbell D, Patton B, Clarke D. Operation for the discrete subvalvular aortic stenosis. J Thorac Cardiovasc Surg 1984;87:366–70. 13. Rodbard S. Physical factors in the progression of stenotic vascular lesions. Circulation 1958;17:410–7. 14. Katz NM, Buckley MJ, Liberthson RR. Discrete membranous subaortic stenosis. Report of 31 patients, review of the literature, and delineation of management. Circulation 1977; 56:1034– 8. 15. Douville EC, Sade RM, Crawford FA Jr, Wiles HB. Subvalvar aortic stenosis: timing of operation. Ann Thorac Surg 1990; 50:29–34. 16. Jacobs JP, Palatianos GM, Cintron JR, Kaiser GA. Transaortic resection of the subaortic membrane: treatment for subvalvular aortic stenosis. Chest 1994;106:46–51. 17. Brown J, Stevens L, Lynch L, et al. Surgery for discrete subvalvular aortic stenosis: actuarial survival, hemodynamic results, and acquired aortic regurgitation. Ann Thorac Surg 1985;40:151–5.
Background. Despite an adequate resection, a significant recurrence rate is encountered in patients undergoing operation for discrete membranous subaortic stenosis. The fibrous membrane and hypertrophied myocardium commonly are removed, but because of the involved risks, the resection may be inadequate and contribute to the recurrence rate. Methods. A review of the cases of 23 patients undergoing operation for discrete membranous subaortic stenosis from 1980 to 1994 was undertaken. Fourteen patients (61%) had coexisting cardiac lesions, all of which were concomitantly repaired. Results. The left ventricle–aorta gradient decreased from a preoperative mean of 63.39 6 7.63 mm Hg to 15.17 6 3.06 mm Hg postoperatively (p < .001) during a
mean follow-up of 3.32 6 0.58 years. Aortic insufficiency decreased postoperatively in 8 patients (34.8%), remained unchanged in 6 patients (26.1%), and showed only insignificant progression in 4 patients (17.4%). There were no early deaths, and the single late death was not cardiac related. No patient had development of endocarditis or heart block or required a pacemaker. One patient (4.3%) had a recurrence, which required reoperation. Conclusions. Our results suggest that aggressive myectomy in concert with membrane resection constitutes safe treatment for discrete membranous subaortic stenosis and is associated with low rates of endocarditis, recurrence, and progression of aortic insufficiency. (Ann Thorac Surg 1997;64:105–9) © 1997 by The Society of Thoracic Surgeons
C
the time of surgical correction for relief of LVOT stenosis, both to prevent the associated sequelae of DMSS and to lessen the likelihood of recurrence. Our results with this technique over a 15-year period are the basis of this report.
ongenital aortic stenosis results from an obstruction of the left ventricular outflow tract (LVOT). The lesion is classified as valvular, subvalvular, or supravalvular, and it can be associated with a variety of other cardiac anomalies. Subvalvular aortic stenosis accounts for 8% to 30% of patients with congenital LVOT obstruction and generally occurs in one of two forms: it occasionally is secondary to a circumferential fibromuscular tunnel, but it more commonly results from a discrete membrane immediately below the aortic valve [1]. The discrete form of subvalvular aortic stenosis (DMSS) has been classified as either a thin, fibrous membrane (type I) or a thicker fibromuscular band (type II) [2]. In our experience, however, all clinically significant lesions have some degree of obstructing hypertrophic myocardium of the LVOT. The natural history of DMSS is one of progressive obstruction, and it is associated with the sequelae of aortic insufficiency (AI), endocarditis, and recurrent obstruction after resection. A major factor in recurrent LVOT obstruction is believed to be inadequate relief of the obstruction at operation. Although the fibrous membrane is universally excised, some groups [3–5] advocate concomitant selective myectomy to achieve full relief of the LVOT stenosis, whereas others [6] have reported that myectomy adds little to the procedure. We believe that aggressive myectomy is necessary at Accepted for publication Jan 14, 1997. Address reprint requests to Dr Heath, Division of Cardiothoracic Surgery, University of Mississippi Medical Center, 2500 N State St, Jackson, MS 39216.
© 1997 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Material and Methods Data Collection All patients undergoing surgical treatment of DMSS from 1980 to 1994 were included in this study; all patients seen with tunnel subvalvular aortic stenosis (n 5 4) were excluded. The case histories of all patients, including patient and primary physician interviews and medical records, were examined retrospectively. All patients were evaluated preoperatively and followed for variable periods postoperatively at The University of Mississippi Medical Center. No patient was lost to follow-up. Information compiled included the patient’s age, sex, previous cardiac operation, and coexisting cardiac anomalies. In addition to the recording of physical signs and symptoms, electrocardiography, chest roentgenography, echocardiography, and often cardiac catheterization were performed preoperatively and at regular postoperative intervals. All complications, reoperations, and deaths were noted. Statistically, left ventricle–aorta gradient measurements collected from cardiac catheterization and echocardiography were used interchangeably and have been shown in prior studies to correlate well [7].
Surgical Procedure All operations were performed by the same surgeon (B.J.H.) at The University of Mississippi Medical Center under stan0003-4975/97/$17.00 PII S0003-4975(97)00371-8
106
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
dard cardiopulmonary bypass, using cold potassium cardioplegia for myocardial protection. Indications for surgical treatment varied but included left ventricle–aorta gradient greater than 30 to 40 mm Hg, evidence of new or progressive AI, and coexisting cardiac lesions requiring operation. A transverse aortotomy was used; the incision was often carried obliquely into the noncoronary sinus. Appropriate traction on the aortic leaflets provided excellent exposure of the subaortic region. The obstructing membrane was circumferentially excised, and a deep, generous trough of hypertrophied muscle in the ventricular septum between the right and left coronary aortic cusps was resected. Great care was exercised to avoid injury to the conduction tissue between the right and noncoronary cusps and the anterior leaflet of the mitral valve, to which the membrane occasionally was adherent.
Statistical Analysis Results are expressed as a range with the mean 6 the standard error of the mean. Group means were compared with Student’s t test for paired events, and event frequencies with a p value of less than 0.05 were considered significant.
Results Preoperative Data From April 1980 to September 1994, 23 patients with DMSS underwent surgical correction. The group comprised 12 female and 11 male patients ranging in age from 7 months to 51 years (mean age, 13.3 6 2.46 years). Five patients (21.7%) had had a previous cardiac surgical procedure: closure of a patent ductus arteriosus in 1; a closed valvulotomy for critical aortic stenosis as a newborn in 1; repair of aortic coarctation in 1; pulmonary artery banding for type I double-outlet right ventricle, closure of a patent ductus arteriosus, and aortic coarctation repair in 1; and repair of an atrioventricular canal defect with mitral valve replacement in 1. Seven patients (30%) were noted to be symptomatic: 3 had dyspnea on exertion (2 with evidence of mild congestive heart failure); 1 had angina pectoris; 2 experienced both dyspnea on exertion and angina; and 1 patient, previously asymptomatic, was seen in ventricular tachycardia after a syncopal episode. The remaining 16 patients (70%) were asymptomatic, and the diagnosis was made on discovery of their murmurs. No patient had endocarditis preoperatively. On physical examination, all patients were noted to have systolic ejection murmurs, 11 with palpable thrills, and 8 had diastolic murmurs of AI. Preoperative evaluation with echocardiography or cardiac catheterization disclosed left ventricle–aorta gradients ranging from 13 mm Hg to 155 mm Hg (mean gradient, 63.39 6 7.63 mm Hg). Fourteen patients (61%) had documented AI preoperatively (11 in 10 patients, 21 in 2 patients, and 31 in 2 adult patients). Of the 23 patients, 9 (39%) had isolated DMSS, and 14 (61%) had coexisting cardiac lesions, all of which were corrected concomitantly at the time of operation for DMSS. One patient undergoing resection of isolated DMSS had had a previous cardiac operation to repair an
Ann Thorac Surg 1997;64:105–9
atrioventricular canal defect with mitral valve replacement and correction of aortic coarctation. Associated lesions and operations included the following: aortic stenosis requiring commissurotomy in 5 patients, 3 of whom had bicuspid aortic valves; aortic valve replacement in 2, 1 of whom had a dysplastic aortic valve and 1, severe aortic regurgitation that was also associated with a patent ductus arteriosus; closure of a ventricular septal defect in 3 patients, one of which was associated with a patent foramen ovale and one, with right ventricular outflow tract obstruction; right ventricular outflow tract obstruction in 1 patient; right ventricular outflow tract obstruction and a patent ductus arteriosus in 1 patient; patent foramen ovale in 1 patient; and type I doubleoutlet right ventricle in 1 patient. Table 1 summarizes the preoperative and postoperative patient data.
Postoperative Findings Postoperative follow-up was possible in all patients and ranged from 7 months to more than 9 years (mean follow-up, 3.32 6 0.58 years). Nonfatal postoperative complications in 5 patients (21.7%) proved transient. These complications included nonspecific perioperative arrhythmias in 3 and high-output renal failure in 2 adults. There were no early postoperative deaths, and to date, there has been only one late death (4.3%), which followed an episode of acute pancreatitis. The postoperative left ventricle–aorta gradient was significantly decreased at late follow-up (p , 0.001) and ranged from 0 mm Hg to 40 mm Hg (mean gradient, 15.17 6 3.06 mm Hg) (Fig 1). Aortic insufficiency progressed in 4 patients (17.4%) (0 to 11), decreased in 8 patients (34.8%) (11 to 0 in 6 patients, 31 to 21 in 1 patient, and 31 to 0 in 1 patient), and remained unchanged in 6 patients (26%); 5 patients had no evidence of AI throughout the study period. Symptoms were completely relieved in all symptomatic patients (71.4%). After resection, 1 patient had intermittent dyspnea on exertion, and another patient, who was seen initially with ventricular tachycardia, had intermittent arrhythmias. Both were treated medically and are currently asymptomatic. Of the 16 asymptomatic patients, 15 remained without symptoms throughout follow-up. One became symptomatic more than 3 years postoperatively, and her symptoms were relieved after reoperation. Two patients required reoperation: the patient just mentioned, who had a morphologically identical recurrence of DMSS, and 1 patient who required closure of a recurrent ventricular septal defect, which was previously repaired at the time of DMSS resection. No patient had progression to the more severe tunnel form of subaortic stenosis, had development of heart block, required a pacemaker, or had development of endocarditis during the study period.
Comment In 1842, Cheevers [8] recorded the first description of subvalvular aortic stenosis. The pathogenesis of this lesion has proved elusive, but a number of theories have
Ann Thorac Surg 1997;64:105–9
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
107
Table 1. Summary of Patient Data Preoperative Data
Age
LV–Ao Gradient (mm Hg)
1
25 y 8 mo
90
2 3 4 5 6 7 8 9 10 11 12 13
51 y 9 y 3 mo 23 y 11 y 32 y 5y 26 y 16 y 13 y 18 y 5 y 6 mo 2 y 8 mo
40 155 70 130 80 70 50 35 60 110 20 80
14 15 16 17
6 y 10 mo 8y 10 y 3 y 8 mo
30 50 40 13
18 19 20 21 22 23
7 mo 4 y 5 mo 1 y 11 mo 5y 14 y 14 y
90 30 90 20 35 60
Patient No.
AI
Associated Procedures
21 PDA ligation, resection RVOT stenosis 0 None 0 None 31 PDA ligation, AVR 0 None 11 Aortic valvotomy 0 PFO closure 11 Aortic valvotomy 11 None 11 None 11 AVR 0 Resection RVOT stenosis 0 DORV repair, PA band removal 0 None 11 VSD closure 11 None 11 VSD closure, resection RVOT stenosis 31 Aortic commissurotomy 21 Aortic commissurotomy 0 Aortic valvotomy 11 PFO and VSD closure 0 None 11 None
Postoperative Data LV–Ao Gradient at Late Follow-up Follow-up (mm Hg)
AI
Endocarditis
2 y 2 mo
30
21
No
No
1 y 6 mo 8y 9 y 9 mo 8 y 1 mo 4y 8 y 10 mo 3 y 1 mo 2 y 5 mo 2 y 10 mo 3 y 6 mo 4 y 10 m 4 y 6 mo
15 26 16 35 25 24 0 0 0 40 0 0
0 0 0 11 11 11 11 0 0 0 11 11
No No No No No No No No No No No No
No No No No No No No No No No No Yes (recurrent DMSS)
1 y 6 mo 1 y 5 mo 7 mo 1 y 5 mo
0 21 0 0
0 0 11 0
No No No No
No No No
1 y 10 mo 1 y 4 mo 1 y 2 mo 1 y 8 mo 1 y 3 mo 8 mo
30 26 25 0 0 36
21 21 0 0 0 11
No No No No No No
No No No Yes (recurrent VSD) No No
Reoperation
AI 5 aortic insufficiency; AVR 5 aortic valve replacement; DMSS 5 discrete membranous subaortic stenosis; DORV 5 double-outlet right ventricle; LV–Ao 5 left ventricle–aorta; PA 5 pulmonary artery; PDA 5 patent ductus arteriosus; PFO 5 patent foramen ovale; RVOT 5 right ventricular outflow tract; VSD 5 ventricular septal defect.
been proposed to explain its occurrence. The explanations attribute DMSS to congenital [9], inflammatory [8], genetic [10], and acquired causes [11]. On the basis of
Fig 1. Change in left ventricle–aorta (LV-AO) peak gradient from before to after operation.
observations of postoperative recurrence identical to the primary lesion, it has been proposed that the nidus is myocardial and that membrane development is a secondary event [12]. We believe the primary lesion is myopathic, which would account for the cases of DMSS in patients with concomitant right ventricular outflow tract stenosis. The progressive nature of DMSS is readily explained by the Rodbard phenomenon [13]. Rodbard reproduced abnormal flow hemodynamics in an animal model, and in accordance with Bernoulli’s principle of energy conservation, progressively stenotic lesions were produced. This would also appear to account for the lesions associated with other defects that produce abnormal flow patterns, such as atrioventricular canal, ventricular septal defect, and double-outlet right ventricle. No one explanation, however, appears to account for all cases of DMSS, and it seems likely to have multiple potential sources. Discrete subaortic stenosis represents a fixed obstruction in which the LVOT stenosis is caused by a specific anatomic lesion and remains unchanged with b agonists. This is in contrast to the dynamic obstruction of idio-
108
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
Ann Thorac Surg 1997;64:105–9
Table 2. Outcome Data From Reported Series
Reference Katz et al (1977) [14] Wright et al (1983) [5] Cain et al (1984) [12] Hardesty et al (1984) [3] Ashraf et al (1985) [6] Brown et al (1985) [17] Douville et al (1990) [15] Jacobs et al (1994) [16] Rayburn et al (1997) [this report]
Early and Late Deaths (%)
AI (% preop)
Recurrence Requiring Reoperation (%)
Endocarditis (%)
7.7 17 2.2 6 8.2 9.4 3 5.6 4.3
29 55 ? ? 18.4 79 59 ? 61
7.7 14 30 14 15 5.7 9 16.7 4.3
13 12 ? 2.9 2 9 3 ? 0
AI 5 aortic insufficiency.
pathic hypertrophic subaortic stenosis, in which the obstruction is secondary to generalized myocardial hypertrophy of the LVOT and worsens with b agonists. Although these have generally been regarded as separate entities, there appears to be a major interrelationship between the two myopathies. A dynamic component is postulated to be a factor in the residual LVOT stenosis of DMSS after adequate resection and may predispose to recurrent obstruction or progression to the more severe tunnellike stenosis [11, 14]. Although knowledge of the pathogenesis and pathophysiology of the lesion is limited, several aspects of the disease are more clearly defined. There is an established association of DMSS with AI and endocarditis, and there is a substantial recurrence rate after surgical excision. Many of the studies in the recent literature include patients who had operations prior to the routine use of cardioplegia for myocardial protection and echocardiography for follow-up, which allow a more complete anatomic resection and noninvasive detection of postoperative sequelae, respectively. Thus, the usefulness of the data in these studies and their conclusions is limited. Our study confirms several aspects of the natural history of DMSS. All patients with marked DMSS have a prominent systolic murmur, often with a palpable thrill and occasionally in concert with a diastolic murmur of AI. The majority of our patients were asymptomatic (70%) but this varies from study to study [14 –17]. We affirm that DMSS often coexists with a variety of congenital cardiac abnormalities; only 39% of our patients had isolated DMSS. Several complications can result from the surgical treatment of DMSS. Overzealous resection in the ventricular septum can lead to heart block from damage to the conduction system, iatrogenic ventricular septal defect, or damage to the anterior leaflet of the mitral valve. Our data confirm that operative treatment of DMSS with extensive myectomy is safe and effective; we had no permanent complications (including development of heart block or requirement of a pacemaker), no early postoperative deaths, and only one late death. That single death occurred more than 6 years postoperatively and was due to complications of acute pancreatitis in an adult patient with multiple medical problems.
The majority of patients with DMSS have abnormalities of the aortic valve, some of which are congenital, but the majority of which are acquired [11]. Aortic insufficiency is the most common acquired lesion of the aortic valve in DMSS and is believed to result from trauma to the valve leaflets by the abnormal jet flow pattern caused by the lesion [5]. The AI can progress postoperatively despite relief of the LVOT stenosis of DMSS [6], but it has been noted that worsening of the AI in DMSS can be slowed or stopped with adequate operative resection of DMSS [5]. Our data support these conclusions. Aortic insufficiency did increase in 17.4% of our patients, but in all of them, the increase was clinically insignificant (0 to 11). In 34.8% of patients with AI preoperatively, the severity decreased. Two of these 8 patients, however, required aortic valve replacement, 1 for a congenitally dysplastic valve and 1 for 31 acquired AI. The aortic valve always remains a potential site for development of endocarditis in patients with DMSS as a result of valve thickening from the jet flow pattern [11], with a reported incidence of up to 13% [14]. The predisposing factors are high LVOT gradient and AI. As with AI, successful relief of LVOT obstruction is believed to reduce the risk of endocarditis in DMSS [5]. None of our patients had development of endocarditis preoperatively or postoperatively during the course of this study. Recurrent LVOT obstruction after resection is well documented, and reoperation is required at a rate of 6% to 30% in similar series [5, 12, 15–17]. This high rate is credited mainly to inadequate resection at the first operation, but a component of dynamic obstruction is also thought to be responsible [11]. The classification of DMSS lesions by Kelly and associates [2] (type I, thin membrane alone, and type II, membrane with associated hypertrophied myocardium) is somewhat misleading in that the lesion of DMSS is progressive and represents but a point in a spectrum in which tunnel stenosis represents the end point. No patient in our series had a membrane without a substantial degree of myocardial hypertrophy obstructing the LVOT (type II), and although reports of type I DMSS are documented in the literature, we believe they are rarely the cause of major LVOT stenosis. Therefore, both myectomy and membrane resection must be undertaken to achieve full relief of the stenosis. To our
Ann Thorac Surg 1997;64:105–9
knowledge, this concept has never been tested in a prospective, randomized trial, and all conclusions are based on retrospective studies; this is likely the result of the relatively infrequent occurrence of this disease. In the series of Ashraf and colleagues [6] and Cain and coworkers [12], use of myectomy did not significantly decrease the number of recurrences that required reoperation. However, both series report reoperation rates in excess of 15%. Our rate of reoperation for recurrence was lower (4.3%) than those previously reported, and no patient had progression to the more severe form of tunnel stenosis. We think that this and our low incidences of AI progression and endocarditis is a result of our commitment to perform an extensive myectomy to fully relieve the LVOT obstruction (Table 2). Resection of the obstructing membrane in concert with generous myectomy of the LVOT is a safe and effective treatment of DMSS. Our results suggest a significant decrease in postoperative left ventricle–aorta gradient, a low rate of recurrent postoperative LVOT obstruction, insignificant progression of AI, and nonexistent rates of endocarditis development and progression to the more severe form of tunnel subaortic stenosis.
References 1. Kirklin JW, Barratt-Boyes BG. Cardiac surgery. 2nd ed. New York: Churchill Livingstone, 1993:1212–24. 2. Kelly DT, Wulfsberg E, Rowe RD. Discrete subaortic stenosis. Circulation 1972;46:309–22. 3. Hardesty RL, Griffith BP, Mathews RA, et al. Discrete subvalvular aortic stenosis. An evaluation of operative therapy. J Thorac Cardiovasc Surg 1984;118:79– 83. 4. McKay R, Ross DN. Technique for the relief of discrete subaortic stenosis. J Thorac Cardiovasc Surg 1982;84:917–20.
RAYBURN ET AL MYECTOMY IN DISCRETE SUBAORTIC STENOSIS
109
5. Wright GB, Keane JF, Nadas AS, Bernhard WF, Castan˜eda AR. Fixed subaortic stenosis in the young: medical and surgical course in 83 patients. Am J Cardiol 1983;52:830–5. 6. Ashraf H, Cotroneo MD, Dhar N, et al. Long-term results after excision of fixed subaortic stenosis. J Thorac Cardiovasc Surg 1985;90:864–71. 7. Lima CO, Sahn DJ, Valdes-Cruz LM, et al. Prediction of the severity of left ventricular outflow tract obstruction by quantitative two-dimensional echocardiographic Doppler studies. Circulation 1983;68:348–54. 8. Cheevers N. Observations on the diseases of the orifice and valves of the aorta. Guys Hosp Rep 1842:387– 442. 9. Keith A. Schorstein Lecture on the fate of the bulbis cordis in the human heart. Lancet 1924;2:1267–72. 10. Pyle RL, Patterson DF, Chacko S. The genetics and pathology of discrete subaortic stenosis in the Newfoundland dog. Am Heart J 1976;3:324–34. 11. Somerville J, Stone S, Ross D. Fate of patients with fixed subaortic stenosis after surgical removal. Br Heart J 1980;43: 629– 47. 12. Cain T, Campbell D, Patton B, Clarke D. Operation for the discrete subvalvular aortic stenosis. J Thorac Cardiovasc Surg 1984;87:366–70. 13. Rodbard S. Physical factors in the progression of stenotic vascular lesions. Circulation 1958;17:410–7. 14. Katz NM, Buckley MJ, Liberthson RR. Discrete membranous subaortic stenosis. Report of 31 patients, review of the literature, and delineation of management. Circulation 1977; 56:1034– 8. 15. Douville EC, Sade RM, Crawford FA Jr, Wiles HB. Subvalvar aortic stenosis: timing of operation. Ann Thorac Surg 1990; 50:29–34. 16. Jacobs JP, Palatianos GM, Cintron JR, Kaiser GA. Transaortic resection of the subaortic membrane: treatment for subvalvular aortic stenosis. Chest 1994;106:46–51. 17. Brown J, Stevens L, Lynch L, et al. Surgery for discrete subvalvular aortic stenosis: actuarial survival, hemodynamic results, and acquired aortic regurgitation. Ann Thorac Surg 1985;40:151–5.