Discrete subaortic stenosis: an acquired heart disease

Abstract

Objective: To determine the anatomic variables in the left ventricular outflow tract in patients with subaortic stenosis. Methods: Between 1982 and 1996, 36 patients were operated on with the `discrete' form of subaortic stenosis (DSS). The mean time of follow up was 7.4 years with a range of 4 months–14 years. There were 25 male and 11 female patients. Mean age at operation was 7.1 years with a range of 9 months–47 years. Results: At the time of surgery, the mitral valve apparatus and interventricular septum were found to be rotated 60–90° in a counterclockwise fashion with anterior displacement into the left ventricular outflow tract in 30 (83%) patients. Subaortic ridge resection with a deep septal myectomy was performed in 32 patients and the remaining four patients had subaortic ridge resection alone. The reoperation free rate at 5 and 10 years were 74±9% and 60±12%, respectively. Reoperations for recurrent disease were performed in 10 (27.7%) patients. No operative or late follow up deaths were encountered. Conclusion: We conclude that DSS is an acquired disease due to a pre-existing anatomic alteration in the mitral valve apparatus and interventricular septum. In addition, recurrence rates are high and physicians should not be mislead by the benign nomenclature its name implies.

Subaortic stenosis, Left ventricular outflow tract

Introduction

Discrete subaortic stenosis is classified as a well-defined membranous or fibromembranous ring, partially or completely encircling the subvalvular aortic region [1]. Proposed etiologies for DSS range from polygenic inheritance [2] to abnormal flow patterns within the left ventricular vestibule [3],[4],[5]. Operative correction often requires sharp ring resection with or without septal myectomy. DSS is not as benign as its name implies. Often DSS is progressive in nature, causing severe left ventricular hypertrophy (LVH) [6],[7] and significant aortic insufficiency (AI) [6],[7],[8],[9]. In addition, recurrence rates are high in several series following the initial resection [9],[10],[11],[12],[13]. The purpose of this paper is to emphasize that DSS is acquired and secondary to pre-existing anatomic alterations in the left ventricular vestibule. In addition, DSS is a progressive disease with frequent recurrences. A patient population closely followed by the same surgical team with a consistent approach to this problem over the past 14 years has enabled us to review these patients with meaningful late results.

Methods

A retrospective review of medical records was performed for DSS at the Oregon Health Sciences University between the years 1982 and 1996. All pertinent data were collected, including sex, date of birth, date of operation, as well as results of echocardiographic and cardiac catheterization examinations before and after operation. Criteria for inclusion in the study were a peak LVOT gradient ≧25 mmHg with progression of LVH or new AI regardless of the gradient. Patients having previous cardiac surgery for congenital heart defects, going on to develop isolated subaortic stenosis were included since the subaortic anatomy is altered artificially in these patients. Six such patients helped us better understand the geometric alterations in the left ventricular vestibule. Conversely, patients with tunnel stenosis, hypertrophic cardiomyopathy and those who had concurrent congenital cardiovascular anomalies repaired at the initial operation for DSS were excluded to isolate a pristine group with the `discrete' subaortic ring anomaly.

Operations were performed with conventional cardiopulmonary bypass at 26°C and cold blood potassium cardioplegia. A transverse aortotomy was carried out and the aortic valve and subvalvular area were exposed. Particular attention was given to documentation of anatomic changes in the subaortic area at the time of operation.

Follow up data were obtained on all patients by telephone contact and complete examination by the primary care physician and/or pediatric cardiologist. Patients' symptomatic states were carefully documented. In addition, recent echocardiographic data were used to document progression or recurrence of disease.

Statistical analysis

Statistical analysis was performed and relevant measures expressed as mean±SE unless otherwise indicated. The Kaplan–Meier product limit method was used to plot the reoperation free curve.

Results

Patient characteristics (Table 1)

Table 1

Patient characteristics

Using the above criteria, 36 patients were identified. Twenty-five (69.5%) were male and 11 (30.5%) were female for a male to female ratio of 2.3:1. The mean age at the time of initial DSS operation was 7.1 years, with an age range of 9 months–47 years (Fig. 1 ). Twelve patients (33.3%) had one or more congenital cardiovascular anomalies previously repaired. These included five total atrioventricular canal repairs, five aortic coarctation repairs, three patent ductus arteriosus (PDA) ligations, one tetralogy of Fallot repair (TOF), one hypoplastic arch repair, and one ventricular septal defect repair (VSD).

Fig. 1

Distribution by age of 36 patients who underwent discrete subaortic stenosis resection.

At the initial DSS operation, 25 (69%) patients were asymptomatic and 11 (31%) were symptomatic. The most common symptom was dyspnea.

Pre-operative hemodynamic results

The mean peak systolic LVOT gradient obtained by either echocardiogram or cardiac catheterization was 58.3±5.05 mmHg (Table 1). Echocardiographic LVH data was available in 31 (86%) patients. Twenty-two (71%) patients had no detectable LVH. The remaining 9 (29%) with LVH were classified as mild in six, moderate in two, and severe in one. Echocardiographic data on AI were available in all patients. Fifteen (41.6%) patients had no identifiable AI. The remaining 21 (58.4%) patients were classified as mild in 20 patients and moderate in one.

Operative procedure at presentation

A discrete subaortic ridge was resected in all 36 patients (Fig. 2 a,b). Ridge resection alone was performed in four patients when the LVOT was quite large. The remaining 32 (88.8%) had ridge resection with a deep septal myectomy (Fig. 3 a,b). In 30 (83%) patients, the surgeon noted that the mitral valve apparatus and interventricular septum had rotated 60–90° in a counterclockwise fashion with anterior displacement into the LVOT. The degree of rotation was determined by visual inspection using the right coronary ostium as a reference point. An echocardiographic comparison of the LVOT between a normal and subaortic stenosis patient is depicted in Fig. 4Fig. 5 . The remaining six (17%) patients who had a previous AV canal or VSD repair were found to have high anterior extension of the VSD Dacron patch closure narrowing the LVOT. There were no aortic root enlargement procedures performed at the initial operation.

Fig. 2

(a) Anatomic appearance of the subaortic ridge. (b) Sharp excision of the subaortic ridge.

Fig. 3

(a) Hypertrophy of the interventricular septum and performance of a deep septal myectomy. (b) Completely excised subaortic ridge with a deep septal myectomy.

Fig. 4

(a) Parasternal short axis view of the left ventricle (LV) showing normal mitral valve (MV) orientation, arrows at medial and lateral commissures. (b) Mitral valve (MV) in patient with subaortic stenosis. Note arrows showing medial commissure rotated anteriorly.

Fig. 5

(a) Parasternal long axis view showing normal relationship of anterior mitral valve (MV) leaflet to left ventricular outflow tract. (b) Patient with muscular subaortic stenosis (SUB AS) showing anterior rotation of mitral valve (MV) leaflet and its proximity to the subaortic stenosis.

Post-operative course and follow up

There were no operative or long-term follow up deaths. The mean follow up time was 7.4 years with a range of 4 months–21 years. Only one patient had post-operative complications which was a 3-year-old child who pre-operatively had a cardiac arrest, prolonged cardiopulmonary resuscitation and subsequent surgery as a salvage procedure. As expected, the patient's hospital stay was prolonged (51 days). The mean intensive care and hospital stay for the remaining 35 patients was 7.6 days with a range of 4–18 days. Early post-operative LVOT gradients were measured by echocardiogram in 29 (80.5%) patients. The mean LVOT gradient was 20.6±2.6 mmHg with a range of 4–58 mmHg.

Long-term mean peak systolic LVOT gradient data measured by echocardiogram were available in 34 (94.4%) patients. The mean gradient was 25.6±3.2 mmHg with a range of 0–64 mmHg. A correlation between the pre-operative, early post-operative and late post-operative mean peak systolic LVOT gradients is depicted in Fig. 6 . Only three patients in long-term follow up had peak systolic LVOT gradients greater than 45 mmHg and are being followed closely for signs of LVH and/or signs of aortic valve dysfunction.

Fig. 6

Correlation between the pre-operative, early post-operative, and late post-operative mean peak systolic LVOT gradients. The LVOT gradients are expressed as mean±SE.

Surgical procedures at reoperation

Reoperation for DSS occurred in 10 (27.7%) patients. The pre-operative mean peak systolic LVOT gradient was 60.7 mmHg. Four patients underwent repeat ridge resection with myectomy, two underwent Konno–Rastan procedures, one underwent posterior root enlargement, and the remaining three had a ridge resection with either myectomy and/or mitral or aortic valve repair (Table 2 ). A third operation, aortic homograft root replacement, was performed in one patient following two prior DSS resections. The reoperation-free states at 5 years and 10 years were 74±9% and 60±12%, respectively (Fig. 7 ).

Table 2

Reoperations

Fig. 7

The reoperative free rate in long-term follow up. AOV, aortic valve; LA, left atrium; LV, left ventricle; LVOT, left ventricular outflow tract; MV, mitral valve; RV, right ventricle.

No reoperative mortalities were encountered. Post-operative complications occurred in two patients both of which were third degree AV blocks necessitating permanent pacemaker implantation.

Discussion

The term LVOT is actually a misnomer. The left ventricle is unable to be anatomically divided into two discrete parts and no true outflow tract is present [14]. However, a narrowed outlet is present and, more properly, is named a vestibule [14]. Since the term LVOT is internationally accepted in current medical nomenclature, it was referred to earlier in this paper.

In this series, the mitral valve apparatus and interventricular septum rotated 60–90° in a counterclockwise fashion with anterior displacement into the left ventricular vestibule in all patients operated on for subaortic stenosis except in those patients who had a prior AV canal or VSD repair. In reference to the latter anomalies, the VSD Dacron patch closure extended high and anterior into the left ventricular vestibule. Operative intervention with resection of the subaortic ring and septal myectomy attempts to enlarge the diameter of this very dynamic area, but does nothing to correct the pre-existing altered geometry of the vestibule. This underlying and inherent geometric problem sets the stage for recurrent disease. Since time of recurrence is inversely related to the extent of subvalvular resection, excision of the fibromembranous ring with a deep septal myectomy offers a lower risk of reoperation than ridge resection alone [9],[10],[15]. Extensive enucleation as described by Ross [16] was not considered, but our policy over the past 14 years has been to perform a myectomy routinely as an adjunct to ridge resection alone.

In our study, no evidence was found to support that DSS was congenital in origin. DSS was not encountered in the newborn period. The youngest patient diagnosed with DSS was 9 months old and the remaining 35 (97.2%) were diagnosed beyond the first year of age. Since only one patient was diagnosed with DSS within the first year, it is unlikely that the disease is truly congenital. Also, in all reoperations for recurrence, the visual and anatomical configuration of the subvalvular ring appeared as if no prior surgery was performed. In each case there was a circumferential thick ridge or membrane very similar to the findings at the original operation. This reinforces our contention that the subvalvular ridge develops over a period of time secondary to a pre-existing anatomic alteration in the left ventricular vestibule.

The unavoidable high rate of recurrence when these patients are followed for many years contradicts the benign perception the term `discrete subaortic stenosis' implies. There is more to the disease than just a thin membrane that can be easily excised. The rate of recurrence requiring reoperation in the literature is as high as 27% [11]. It is for this reason that we recommend extensive myectomy at the time of initial surgery to provide a longer symptom- and disease-free state. When a second operation becomes necessary, patients usually have grown in size allowing an even more extensive operation, if necessary, such as a limited or extensive aortic root enlargement. In our series, three patients eventually underwent either a posterior root enlargement or a Konno–Rastan type aortoventriculoplasty due to failure of the aortic annulus to grow. This again demonstrates the multiple anatomic factors including the aortic valve and annulus, mitral valve and subvalvular mitral valve apparatus, and the interventricular septum that account for the varying speed of disease progression and age of presentation.

Surgery for DSS is safe and complications like aortic or mitral valve injury, complete heart block, and inadvertent creation of a VSD are quite rare [17]. In our series there were no early or late deaths and only two post-operative complications which were in the reoperative population.

Conclusion

Discrete subaortic stenosis is a manifestation of a geometric anatomic alteration in the left ventricular vestibule. Patients who have surgical intervention for discrete subaortic stenosis need to be followed closely for the rest of their lives. Recurrence should be addressed promptly with the same indications as at the time of initial surgery. In addition, physicians involved in the care of these patients should not be mislead by the benign nomenclature its name implies.

© Elsevier Science B.V. 1998