Discrete Subaortic Stenosis

Abstract

Discrete subaortic stenosis, which is an obstructing lesion of the left ventricular outflow tract, remains a surgical challenge. The recurrence rate is high despite sufficient conventional resection.

We retrospectively reviewed the results of surgery for discrete subaortic stenosis at our institution from September 1995 through March 2001. Twenty-one patients with this lesion underwent surgical treatment during this period. Excision of the fibromuscular membrane with myectomy was performed in all of the patients. Follow-up in all patients ranged from 7 to 67 months (mean follow-up period, 39.57 ± 15.46 months).

The mean systolic gradient between the left ventricle and the aorta decreased from 59.23 ± 35.38 mmHg preoperatively to 9.47 ± 9.91 mmHg postoperatively. There was no instance of heart block that required a permanent pacemaker, nor of bacterial endocarditis. There was no early or late postoperative death. A 22nd patient, who had 3+ aortic regurgitation, required aortic valve replacement and was excluded from the study. Two of the patients (9.5%) underwent reoperation because of recurrent gradient and residual ventricular septal defect.

Our results suggest that fibromuscular membrane excision combined with myectomy in patients with discrete subaortic stenosis produces sufficient relief of obstruction with low morbidity. (Tex Heart Inst J 2003;30:286–92)

Congenital obstruction of the left ventricular outflow tract (LVOT) affects approximately 3% to 10% of individuals with congenital heart disease, and discrete subaortic membrane accounts for 8% to 10% of all cases of LVOT obstruction in children. ^1–3 This last condition presents as a membranous or fibromuscular ring below the aortic valve, either in isolation or in association with other congenital anomalies such as ventricular septal defect, patent ductus arteriosus, coarctation of the aorta, bicuspid aortic valve, abnormal left ventricular (LV) papillary muscle, atrioventricular septal defect, and persistent superior left vena cava. The condition is rarely diagnosed antenatally or in infancy but often manifests in the 1st decade of life with features of progressive LVOT obstruction, LV hypertrophy and dysfunction, or aortic regurgitation. The jet from the narrowed subaortic tract damages the aortic cusps and causes regurgitation; this damage may also render the aortic valve prone to infective endocarditis. ⁴

Treatment is usually surgical. Despite adequate resection, there is a substantial recurrence rate among patients who have undergone operation for discrete membranous subaortic stenosis (DSS). Although membranectomy, with or without septal myotomy or myectomy, has been the accepted method of treating fixed subaortic stenosis, there are still controversies concerning operative methods and uncertainties concerning the recurrence of subaortic obstruction and the development of aortic insufficiency after repair. Our usual approach to DSS is membrane excision with resection of the septal myocardium. In this study, we aimed to evaluate our surgical results.

Patients and Methods

Data Collection. All patients who underwent surgical treatment of DSS at our institution from September 1995 through March 2001 were included in this study. Patients with significant valvular aortic stenosis, tunnel LVOT obstruction, or idiopathic hypertrophic cardiomyopathy were excluded. One patient who underwent aortic valve replacement combined with aortoventriculoplasty was excluded from the study. All patients were evaluated preoperatively and monitored for variable periods postoperatively. All patients or their families signed informed consents for investigation during the study period. The case histories of all patients, including patient and primary physician interviews and medical records, were examined retrospectively. We compiled information on each patient's age, sex, previous cardiac operation or co-existing cardiac anomaly (if any), physical signs and symptoms, electrocardiography, chest radiography, echocardiography, and cardiac catheterization. Echocardiography was performed preoperatively and at regular postoperative intervals (Fig. 1). All complications and reoperations were noted.

Fig. 1 A) Parasternal long-axis view of the left ventricular outflow tract shows a subaortic discrete membrane (arrow). B) Postoperative parasternal long-axis view in the same patient demonstrates relief of the subaortic stenosis.

All echocardiographic and angiocardiographic studies were conducted by the same pediatric cardiologist (EL). The maximal instantaneous gradient across the LVOT was calculated from the peak spectral Doppler velocity using the modified Bernoulli equation. ⁵ Aortic regurgitation was characterized as follows: 0, no aortic regurgitation; 1, trivial (thin regurgitant jet seen at the valve leaflets, no left ventricular distention or diastolic retrograde aortic flow in the descending aorta); 2, mild (slightly broader jet limited to the LVOT, no ventricular enlargement, minimal or no retrograde flow in the descending aorta, pressure half time = 400 ms); 3, moderate (broad regurgitant jet extending into left ventricle, mild ventricular enlargement, holodiastolic retrograde aortic flow in the descending aorta, pressure half time 200–400 ms); and 4, severe (wide regurgitant jet extending deep into the left ventricle, marked left ventricular enlargement, holodiastolic retrograde flow in the descending aorta, pressure half time <200 ms). Cardiac catheterization was per formed in 9 patients in whom no satisfactory images could be obtained on echocardiography. The peak gradient across the left ventricle–aorta was measured by catheter pull-back and by monitoring pressures in both chambers.

Surgical Procedure. Indications for surgery varied but included left ventricle–aorta gradient that exceeded 30 mmHg, coexisting cardiac lesions that required operation, and echocardiographic or angiographic evidence of new or progressive aortic regurgitation. All operations were performed by the same pediatric cardiac surgeon (EAA) at the department of cardiac surgery in Ege University Hospital. All operations were performed with the patient on standard cardiopulmonary bypass under moderate systemic hypothermia. Myocardial protection was achieved by means of cold blood cardioplegia—antegrade, retrograde, or both. After cardiopulmonary bypass was instituted, the aorta was opened by an oblique aortotomy that was extended into the noncoronary sinus. After carefully retracting the leaflets of the aortic valve, surgeons could see the subaortic membrane (Fig. 2). They circumferentially excised both the obstructing discrete membrane and the adjacent hypertrophied muscle. 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 adheres. The mean cardiopulmonary bypass time was 68 ± 41 minutes, and the mean cross-clamp time was 30 ± 17 minutes.

Fig. 2 Intraoperative views of A) subaortic discrete membrane (arrow) and B) left ventricular outflow tract after membranectomy.

Statistical Analysis. Results are expressed as a range, with the mean ± SE. Group means were compared with Student's t-test on SPSS software (SPSS Inc.; Chicago, Ill); a P value of less than 0.05 was considered significant.

Results

Preoperative Data. From September 1995 through March 2001, 21 patients with DSS underwent resection of discrete subaortic stenosis alone or as part of a more comprehensive surgical procedure. The group comprised 9 female and 12 male patients who ranged in age from 27 months to 61 years (mean age, 12.64 ± 16.19 years). Three patients (14.3%) had undergone a previous cardiac surgical procedure: ventricular septal defect (VSD) closure in 1; coexisting VSD closure and right ventricular outflow tract (RVOT) stenosis repair in 1; and coexisting atrial septal defect (ASD) and VSD closure in 1.

Eight patients (38.1%) were noted to be symptomatic: 4 patients showed diminished exercise tolerance, 1 had palpitations, and 1 had syncopal episodes. The remaining 13 patients (61.9%) were asymptomatic. On physical examination, all 21 patients revealed systolic ejection murmurs, and 15 had diastolic murmurs of aortic regurgitation. The left ventricle– aorta gradient ranged from 14 to 155 mmHg (the mean gradient was 59.23 ± 35.38 mmHg). Eleven patients had 1+ aortic regurgitation, and 4 patients had 2+ aortic regurgitation. Two patients had mild mitral insufficiency. No patient had endocarditis preoperatively.

Of the 21 patients, 15 (71.4%) had coexisting cardiac lesions. Of these, 6 patients had a VSD, 3 had RVOT stenosis, 2 had a patent foramen ovale (PFO), 2 had persistent left superior vena cava (LSVC), and 2 had mild mitral insufficiency. All conditions but the persistent LSVC and the mild mitral insufficiency were corrected concomitantly at the time of DDS surgery. The 2 patients with mild mitral insufficiency did not require concomitant surgical correction. The remaining 7 patients (33.3%) had isolated DSS, as summarized in Table I.

TABLE I. Summary of Patient Data

Postoperative Findings. The patients underwent postoperative follow-up that included physical examination and echocardiography. Follow-up results were available for all patients, and length of follow-up ranged from 7 to 67 months (mean, 39.57 ± 15.46 months). Two patients (9.5%) developed nonfatal cardiac arrhythmia during the perioperative period and were treated successfully. One patient (4.8%) developed acute pancreatitis during the postoperative period and was treated medically with complete recovery. Bundle branch block developed in 8 patients during the postoperative period. No patient developed endocarditis or heart block that required a pace-maker. There were no early or late postoperative deaths. The early postoperative echocardiographic results showed that the membrane was completely removed in all patients. At late follow-up, the systolic gradient between the left ventricle and the aorta ranged from 0 to 38 mmHg (mean gradient, 9.47 ± 9.91 mmHg), which was significantly lower than the preoperative values (P <0.001) (Fig. 3). Aortic regurgitation was also evaluated by means of echocardiography during the postoperative period. It decreased in 8 patients (38.1%) and progressed in 4 patients (19%). The aortic competence in the remaining 9 patients (42.9%) was unchanged. Two patients (9.5%) required reoperation: 1 patient had recurrent DSS, and 1 patient required closure of a recurrent ventricular septal defect, which had been repaired at the time of DSS resection.

Fig. 3 Postoperative changes in left ventricle–aorta (LV–Ao) peak gradient.

Symptoms of the patients were relieved completely during the early postoperative period, except for palpitation in 1 patient, which resolved with medical therapy.

Discussion

Discrete subaortic stenosis is a manifestation of a geometric anatomic alteration in the LVOT. This endocardial abnormality involves not only the subaortic ridge but also the leaflets of the adjacent valves. ¹ Although substantial pressure gradient and aortic regurgitation are the main indications for surgery, controversy persists about the timing of surgical repair and the surgical technique. ^3,5

Unlike valvular aortic stenosis, DSS seems to be acquired, because it is almost never present at birth. It is believed that obstruction in this instance is a consequence of some abnormality of ventricular motion, growth, hypertrophy, or a combination of these factors. 1,3,4,6–8 An abnormal angle between the muscular and conical ventricular septum appears to be an important causative factor, but a definitive cause has not been established. ^9–11

Various theories about the cause of DSS—congenital, inflammatory, genetic, and acquired—have been proposed. ^3,12 Some investigators have considered the disease to be a form of cardiomyopathy, ¹³ and there are patients with DSS who develop an unusually profound LV myocardial hypertrophic response. More often, however, there appears to be no global cardiomyopathic component. In the vast majority of cases, LV hypertrophy regresses after relief of the outflow obstruction, as would be expected in a patient with valvular aortic stenosis. We are convinced that DSS has multifactorial causes.

Discrete subaortic stenosis is sometimes associated with various other cardiac malformations 1–4,7,8 that must be monitored and treated surgically when necessary. Chung and colleagues ¹⁴ reviewed the cases of 8 patients, aged 1 to 8 years, who had VSD in combination with various other cardiac lesions, but no evidence of DSS. They operated on 6 of these patients for the associated lesions, but none required closure of the VSD. In 6 of the 8 patients, the VSD closed spontaneously, and in the remaining 2 there was a small residual VSD. Subsequent serial echocardiography showed evidence of a subaortic membrane, which was confirmed by catheterization and then by surgical resection in 5 patients. They concluded that DSS might develop in patients with small or spontaneously closed VSDs. Rayburn and associates ³ recorded 3 cases of VSD in their DSS patients and closed all of the defects. We encountered 6 cases of VSD in our patients who had DSS, and we closed all of the ventricular defects during resection of the fibromuscular membrane. We also resected 3 stenoses of the RVOT, ligated 2 patent ducti arteriosi (PDAs), and closed 2 PFOs. Discrete subaortic stenosis can sometimes require direct procedures to enlarge the LVOT, especially when DSS coexists with the diffuse type of LVOT stenosis. Such a procedure is generally conducted as an aortoventriculoplasty. ^1,15 We performed it in the patient who underwent aortic valve replacement and was subsequently excluded from this study.

Although the most common symptom in patients with DSS is diminished exercise tolerance, they may also have syncope or angina pectoris. ^1,3,4 It should be remembered that most patients are asymptomatic, even in the presence of important gradients that indicate surgery. ^1,3,8 Rayburn and colleagues ³ confirmed that 70% of their patients had no symptoms, but the proportion varies from series to series. Kuralay and coworkers ¹⁶ reported that 64.4% of their patients had exertional dyspnea. In our series, 8 patients (38.1%) were noted to be symptomatic, and the remaining 13 (61.9%) were asymptomatic.

Apart from congenital cardiac abnormalities, acquired aortic insufficiency is the most common lesion found in association with DSS, and it can of course be progressive. The aortic regurgitation appears to be due to thick fibrous tissue on the left ventricular surface of the valve leaflets. The fibrosis is caused by repetitive trauma from a jet of blood through the stenosis or by the proliferation of the fibroelastic membrane itself. This fibrous tissue can play an important role in the retraction of the valve leaflets. ¹⁷ Although an important study found no benefit in early surgery and a higher prevalence of aortic regurgitation in surgically treated patients, ¹⁸ we think that early surgery may preserve the integrity of the aortic valve in such a manner as to avoid later valve replacement. Aortic insufficiency can progress postoperatively despite relief of the LVOT stenosis of DSS, but our operative results and early follow-up tend to support the conclusion that worsening of the regurgitation in DSS can be slowed or stopped with adequate resection of DSS. This has been affirmed by various other studies. 3–5,15,16

Although DSS has been treated surgically for many years, ^1,7 the optimal operative management and the timing of surgery remain controversial. ^5,12 Many authors have suggested surgery for patients who have left ventricle–aorta gradients that exceed 30 mmHg or a coexisting cardiac defect that requires surgical correction, ^1,5,8 while others advocate surgical resection for DSS of any degree because of concerns about the developmental role that subaortic stenosis may play in aortic insufficiency. ^12,18 This earlier intervention for DSS can prevent the development of abnormal muscular hypertrophy and can reduce the occurrence of aortic regurgitation. It also protects patients from bacterial endocarditis. 1–3,5,7

The optimal surgical method for patients with DSS is debatable. ^3,5 Although some surgeons prefer enucleation of the discrete membrane and in selected patients its fibromuscular ridge, ⁸ many others believe that a routine myectomy is adequate to eliminate recurrence. The latter, however, confirm that surgery is not sufficient without resection of hypertrophied muscle. ^5,19 We too resected adjacent hypertrophied muscle during resection of fibrous membrane in all of our patients. Limited septal muscle resection can relieve obstruction of the LVOT in some patients, but a more aggressive approach must be taken when the pathologic condition requires it. Some of the residual perioperative left ventricular–aortic gradients in patients with discrete subaortic stenosis undergoing repairs are dynamic and transient, and are probably related to increased sympathetic activity following surgery. However, persistent dynamic obstruction was found in 44% of patients after removal of discrete subaortic obstruction. ¹³ Several groups have reported the need for repeat myectomy to relieve obstruction after the successful removal of subaortic membrane. Because the fibrous tissue that retracts the aortic valve leaflets remains, there can be no improvement in the aortic regurgitation after surgery. Radical excision of all diseased tissue, which attains a minimal early postoperative gradient, may reduce the occurrence of late aortic regurgitation. ²⁰ However, this more aggressive approach increases the risk of iatrogenic damage to the conduction tissue, ventricular septum (VSD), and mitral valve. Kuralay and coworkers ¹⁶ have reported fewer complications, such as conduction tissue injury, when myectomy is guided by transesophageal echocardiography. We believe that the risk of developing heart block that requires permanent pacing is minimal when the surgeon takes great care to avoid injury to the conduction tissue between the right and noncoronary cusps.

Early results after surgery for DSS are usually good. The left ventricle–aorta gradient was decreased significantly in many studies similar to ours. 3,5,7,8,19,21 Two of our patients had residual gradients of 30 mmHg or greater, but no progression was seen in early follow-up. The failure to eliminate their residual gradients may be due to incomplete resection of the hypertrophied myocardium. In recent series, ^9,11 postoperative aortic regurgitation has decreased substantially. Although the rate of bacterial endocarditis ^22,23 was relatively high in older reports, it too has decreased in recent studies. ³ In most of the series, the rates of peri operative mortality and cardiac-related late death have been low. ^5,15,21 Jahan-giri and associates ¹⁵ affirmed that they had no postoperative early or late deaths in their patients who had undergone surgery for subaortic stenosis. Postoperative complications and recurrences are also low in these patients. The residual gradients that are seen in some patients (including 2 of our own) may be due to insufficient resection of the fibromuscular diaphragm and the coexisting hypertrophied muscle that together cause dynamic outflow obstruction. ^5,8 Complete heart block and bundle branch block can follow extensive muscle resection. ^24,25 In our procedure, we routinely resected hy pertrophied myocardium. We had 1 patient (4.8%) with a residual gradient that required reoperation, and no patient with heart block that required a permanent pacemaker. Eight of our patients (5 of whom had undergone concomitant VSD closure) developed bundle branch block postoperatively. Bacterial endocarditis has been considered an important risk attendant upon DSS, especially in former series. 1,3,5,7,8,22 We encountered no bacterial endocarditis in our patients, as some authors did. ^24,25

Although DSS is a well-known cardiac anomaly that is seen in both children and adults, there is no agreement about its surgical management. 5,8,12,18 As our experience shows, membranectomy together with routine myectomy and (when necessary) a more aggressive approach seems to yield acceptable morbidity, mortality, and recurrence rates, compared with series that have used only membranectomy. However, our follow-up period was relatively short, and our concept has never been tested in a prospective, randomized trial. Due to the infrequent occurrence of this disease, our conclusions are drawn from retro spective study. We must await the long-term results in these same patients before our approach can be justified. In conclusion, we believe that a larger number of patients, longer follow-up by the pediatric cardiologist, and close collaboration with the pediatric cardiac surgeon will be necessary to support this management.