Mid- and long-term outcomes after surgical correction of subaortic stenosis: a 27-year experience

Abstract

Open in new tabDownload slide

OBJECTIVES

We reviewed the mid- and long-term surgical outcomes of patients with subaortic stenosis (SAS).

METHODS

Patients operated for SAS from April 1990 to August 2016 were reviewed retrospectively. Patients with major associations such as aortic arch obstruction were excluded. Time to reintervention and predictors of recurrence were assessed using Kaplan–Meier analysis, log-rank test and uni/multivariable Cox regression.

RESULTS

120 patients at a median age of 4.7 years (interquartile range 2.9, 8.1) underwent primary operation (median peak preoperative left ventricular outflow tract gradient 52.5 mmHg, interquartile range 40, 70) involving fibrous tissue excision (n = 120) with septal myectomy (93%; n = 112) as the procedure of choice.

At median follow-up of 13 years (interquartile range 7, 18), freedom from reintervention at 1, 3, 5 and 10 years was 99% (95% confidence interval 94%, 99%), 94% (87%, 97%), 93% (86%, 96%) and 90% (82%, 94%), respectively. Recurrence occurred in 18% (n = 20) with 15 patients undergoing reinterventions, 13 of whom required radical reoperation. Multivariable analysis revealed higher preoperative peak left ventricular outflow tract gradient (hazard risk 1.06, confidence interval 1.03, 1.09, P < 0.001), and presence of bicuspid aortic valve (hazard risk 14.13, confidence interval 3.32, 60.1, P < 0.001) as predictors for reintervention. Mild/moderate aortic regurgitation occurred in 49% (n = 55) of patients at the most recent follow-up.

CONCLUSIONS

Reintervention for recurrent SAS is common, predicted by higher preoperative peak left ventricular outflow tract gradient, and presence of bicuspid aortic valve, and frequently involves a radical procedure. Aortic regurgitation is a major consequence of SAS, but its severity usually remains low.

Clinical Registration Number

SCHN HREC reference number 2019/ETH02729, approved on 09 July 2019.

INTRODUCTION

Subaortic stenosis (SAS) represents a rare congenital heart defect accounting for 1–2% of all congenital cardiac anomalies [1]. Discrete SAS occurs in about 6% of children with congenital heart defects [2–4] and is responsible for 8–30% of left ventricular tract obstruction (LVOTO) in these individuals [5–8].

Although SAS is classified as a congenital abnormality, familial occurrence is rare with little evidence suggesting that it is a primary genetic disorder [9]. Acquired SAS usually develops within the first decade of life secondary to certain morphological abnormalities in the left ventricular outflow tract (LVOT) such as an acute aorto-septal angle, smaller aortic annulus and large aortic-mitral valve separation distance [10–12]. The resulting fluid-induced stress triggers loss of tissue homeostasis, endothelial cell migration, proliferation and activation of a cascade of signals leading to inflammation and fibrosis in the LVOT [13].

Surgical excision of the fibrous tissue with or without adjunct septal myectomy has been the treatment of choice for SAS with good results [5, 14, 15] but recurrence is common, between 13–31% [1, 5, 12, 15, 16]. Risk factors for recurrence are female gender, preoperative peak LVOT gradient >50 mmHg, younger age at surgery, aortic valve-to-membrane distance <5 mm and residual end-operative peak gradient >10 mmHg [7, 12, 15, 17–20].

The purpose of this study was to review the mid- and long-term outcomes following surgical management of patients with SAS and to define the risk factors for recurrence.

MATERIALS AND METHODS

Ethics statement

The Sydney Children’s Hospitals Network Human Research Ethics Committee approved this study (Reference number 2019/ETH02729, approved on 09 July 2019). A waiver of patient consent was approved by the ethics committee in view of anonymity. The study was conducted at The Children’s Hospital at Westmead, and at The Sydney Children’s Hospital at Randwick.

Patients

Between April 1990 and August 2016, 120 patients underwent surgical repair of SAS at these 2 hospitals in Sydney. Data was collected retrospectively through hospital medical records, echocardiography records and outpatient follow-up letters. This series includes all patients with SAS with or without associated cardiac anomalies (Supplementary Material, Table S1) including ventricular septal defect (VSD), referred for surgical management during this period. Patients with major concomitant cardiac anomalies such as hypoplastic aortic arch, interrupted aortic arch, severe valvular aortic stenosis and hypertrophic obstructive cardiomyopathy were excluded. Patients developing LVOTO following repair of double outlet right ventricle or repair of atrioventricular septal defect were excluded as well.

SAS was classified as discrete (short segment) or tunnel type (long segment) according to its echocardiographic appearance and this is usually an arbitrary estimate [2]. Recurrence of SAS was defined as peak LVOT gradient of ≥40 mmHg at any time after the first postoperative month [9]. Related reoperation was defined as any procedure performed to readdress the recurrence of SAS.

Surgical technique

Trans-aortic excision of the subaortic fibrous ridge was performed circumferentially. Most of the patients underwent adjunct septal myectomy.

Fibrous ridge extension on the aortic valve cusp/s and/or the anterior mitral leaflet was peeled off employing blunt dissection. Every attempt was made to excise the abnormal fibrous tissue in total to avoid recurrence. The adequacy of excision was determined arbitrarily by a visual estimate, sometimes by admitting an appropriately sized Hegar dilator and mostly confirmed by intraoperative transoesophageal echocardiography after weaning from bypass. A residual peak LVOT gradient ≤15 mmHg has been defined as a satisfactory intraoperative result in our institution; however, intraoperative gradients have not routinely been measured across the time period of the study preventing this data from being included.

Reoperation for recurrence of SAS included re-resection of the fibrous ridge/membrane with septal myectomy ± Modified Konno procedure [21] or a Ross–Konno procedure [22] in patients with a small aortic root with multi-level LVOTO. During each echocardiographic assessment, peak flow velocity across the LVOT was measured by pulsed and continuous wave Doppler techniques and then the peak LVOT pressure gradient was calculated using simplified Bernoulli equation [23]. Aortic valve regurgitation was quantified using colour Doppler imaging and graded according to the ratio of width of regurgitant jet to the diameter of LVOT (mild <25%; moderate 25–65%; severe >65%) [24], and by assessing the reversal of flow in descending thoracic aorta.

Statistical analysis

There were 120 patients in the study group but follow-up was incomplete in 7, who were missing all postoperative data, including whether a reintervention occurred and a last follow-up, thus were excluded from the follow-up analysis. One late death was censored at that event. A Kaplan–Meier curve was constructed to illustrate the time to first reintervention for LVOTO among the full sample, and to assess association between bicuspid aortic valve and time to reintervention for LVOTO. Covariables were selected after review of literature and through discussions amongst coauthors. Chosen covariables for adjustment were bicuspid aortic valve, gender, aortic annulus z-score, age at surgery, removal of fibrous ridge from the left ventricular septum and/or aortic cusps and/or anterior mitral leaflet and preoperative peak LVOT gradient.

Secondary analysis for incidence of aortic regurgitation was performed using Kaplan–Meier curve to demonstrate the time to reintervention depending on preoperative class of severity of aortic regurgitation. Univariable Cox regression was also utilized to determine the pre-operative/postoperative predictors of mild/moderate aortic regurgitation (Grade 1+/2+) at latest follow-up.

All analyses were undertaken in R statistical software version 4.2. 1 with statistical significance set at P < 0.05. Data underlying this article are available in the article and in its Supplementary Material.

RESULTS

Patients’ characteristics and cardiac disease morphology

There was male predominance in this study group constituting 59% (n = 71) of the patients. SAS was discrete in 92% (n = 110) of the patients and was tunnel-type in the remaining 8% (n = 10), with a circumferential lesion in 64% (n = 77) of the total cohort. Spectrum of associated lesions and syndromes at the time of presentation have been summarized in Supplementary Material, Table S1. Aortic regurgitation was present in majority of the patients (71.6%; n = 86) but was either trivial (27.5%; n = 33) or mild (44.1%; n = 53) (Fig. 1).

Twelve patients (10%) underwent a pre-repair procedure in the form of either VSD closure (n = 4), and/or balloon dilatation of aortic valve (n = 7) and/or patent ductus arteriosus ligation (n = 2) and/or cor triatriatum repair (n = 1) and/or repair of coarctation of aorta (n = 1). The median peak gradient across the LVOT at the time of surgical repair was 52.5 mmHg [interquartile range (IQR) 40, 70] and the median age was 4.7 years (IQR 2.9, 8.1).

Procedural analysis

Excision of fibrous ridge/membrane was performed in all 120 patients. The fibrous tissue was excised from the interventricular septum (93%; n = 112), from the ventricular side of the anterior mitral leaflet (46%; n = 55) and from the undersurface of the aortic cusps (29%; n = 35) to ensure complete relief. Figure 2 summarizes the involvement of different aortic cusps by the fibrous tissue. Adjunct septal myectomy was performed in 112 patients (93%). Closure of VSD (21.6%; n = 26) was the commonest procedure performed simultaneously with SAS repair (Table 1). The median cardiopulmonary bypass time was 57 mins (IQR 48, 76) and median aortic cross-clamp time was 36 mins (IQR 30, 51).

The procedural complications included occurrence of cardiac arrhythmia in 10 patients (8.3%) in the form of new left bundle branch block (n = 7; 5.8%), new right bundle branch block (n = 2; 1.6%) and complete heart block (n = 1; 0.8%) requiring permanent pacemaker insertion. Iatrogenic VSD occurred in 2 patients (1.6%), both were diagnosed postoperatively, and the shunt was nonsignificant at that time, so were left alone. Immediate postoperative echocardiography (predischarge) study revealed a median residual LVOT peak gradient of 8 mmHg (IQR 0, 15) and presence of aortic regurgitation in 63 patients (52.5%), none more than mild in severity (trivial in 21 patients, 17.5%; and mild in 42 patients, 35%) (Fig. 1). There was no hospital mortality in this series.

One-year outcomes

The median LVOT peak gradient was 8 mmHg (IQR 0, 12) with recurrence of SAS in 3 patients (2.5%) with a mean peak LVOT gradient of 67 mmHg (standard deviation 25 mmHg). Aortic regurgitation was documented in 74 patients (61.6%) of which it was trivial in 17 (14.2%), mild in 55 (45.8%) and moderate in 2 patients (1.6%) (Fig. 1). There was also valvar aortic stenosis noted in 19 patients (15.8%), being mild in 18 of them (15%) and moderate in just 1 (0.8%).

Long-term outcomes

Median follow-up for this study was 13 years (IQR 7, 18). Figure 3 shows the Kaplan–Meier curve of time to LVOTO reintervention among the full sample. Freedom from reintervention at 1, 3, 5 and 10 years was 99% [95% confidence interval (CI) = 94%, 99%], 94% (CI = 87%, 97%), 93% (CI = 86%, 96%) and 90% (CI = 82%, 94%), respectively. Table 2 summarizes the list of reinterventions.

Reintervention group

There were a total of 25 re-interventions including multiple reinterventions on many patients, of which 22 were related to LVOTO. Fifteen patients had one reintervention, 8 had 2nd reintervention, and 2 had 3rd reintervention. All 15 patients who had at least one reintervention during the follow-up period, had undergone fibrous excision + septal myectomy as a primary operation. The median age for primary operation in this group was 3 years (IQR 1, 5), with a range of 4 months to 15 years. The spectrum of procedures at 1st reintervention were fibrous ridge excision + modified Konno procedure (n = 6), fibrous ridge excision only (n = 3), fibrous ridge excision + aortic valve repair (for mild aortic stenosis) (n = 2), Konno procedure + aortic valve replacement (AVR) (n = 2), and fibrous ridge excision + septal myectomy + AVR (n = 1). The median age for 1st reintervention was 9.8 years (IQR 4, 14).

One patient who had a non-related 1st reintervention for iatrogenic VSD closure, had a 2nd reintervention involving fibrous ridge excision + septal myectomy + AVR. Second reintervention was required in 8 patients with Ross–Konno procedure in 3 and AVR in 2. The median age for 2nd reintervention was 12.5 years (IQR 6, 16). One patient required 3rd related reintervention in the form of Ross–Konno procedure. Of the total 4 patients with Ross–Konno procedure in this group, one required catheter-based pulmonary valve replacement for homograft stenosis. A total of 9 patients underwent aortic valve replacement with either a prosthesis (n = 5) or a pulmonary autograft (n = 4).

Table 3 shows the univariable and multivariable associations between the predictors and LVOTO reintervention. Presence of bicuspid aortic valve was associated with >9 times increased risk of experiencing a reintervention [hazard risk (HR) = 9.41, 95% CI (3.09, 28.63), P < 0.001]. Smaller aortic valve annulus z-score, and higher preoperative peak LVOT gradient were also associated with reintervention in the univariable analysis.

After adjustment, presence of a bicuspid aortic valve remained strongly associated with the risk of reintervention [HR = 14.13, 95% CI (3.32, 60.1), P < 0.001], as did a higher pre-operative peak LVOT gradient [HR = 1.06, 95% CI (1.03, 1.09), P < 0.001] (Table 3).

Analysis of association with bicuspid aortic valve

The Kaplan–Meier curve showing the time to reintervention by bicuspid status (Fig. 4) revealed that the increased risk was not constant (i.e. proportional hazards) but confined to the first 2 years following surgery. A scatterplot of pre- and postoperative peak gradient by bicuspid aortic valve (Supplementary Material, Fig. S1) was constructed which demonstrated that peak gradient was greater at both time-points among those with bicuspid valve.

Secondary analysis for aortic regurgitation

A Kaplan–Meier curve of time to reintervention by preoperative class of severity of aortic regurgitation (Supplementary Material, Fig. S2) did not reveal any association with LVOT reintervention. For predictors of aortic regurgitation at the latest follow-up, the analysis was dichotomised as none/trivial versus mild/moderate. The univariable analysis (Table 4) revealed that no covariable predicted occurrence of mild/moderate aortic regurgitation at latest follow-up.

At the end of the follow-up period, the median peak LVOT gradient was 4 mmHg (IQR 0, 15), aortic regurgitation present in 86 patients (76.8%), being trivial in 31 (27.7%), mild in 47 (42%), and moderate in 8 patients (7.1%) (Fig. 1). No patient required reintervention specifically for aortic valve regurgitation during follow-up. Valvar aortic stenosis was noted in 12 patients (10.7%) which was mild in 11 (9.8%) and moderate in one (0.9%).

DISCUSSION

SAS as a cause for LVOTO is a well-known entity in paediatric patients. Isolated SAS with or without associated VSD can be congenital or acquired in origin [13]. The most convincing aetiology is acquired origin of SAS secondary to wall shear stress in the LVOT [25].

This study investigates a large longitudinal cohort of SAS patients with a long follow-up duration. Therefore, the data provided essentially tracks the patient’s unnatural history following surgical intervention for SAS and demonstrates that in some patients there is development of multi-level LVOTO during follow-up that requires more aggressive reintervention. Moreover, necessity for lifelong follow-up after intervention is supported by the need for reintervention up to 15 years after the primary procedure.

Similar to the anatomic factors leading to increased wall shear stress in the subaortic region, the morphology of bicuspid aortic valve has been linked to the existence of altered haemodynamics near the aortic valve leaflets, leading to valvular calcification and aortic medial degeneration [26]. But bicuspid aortic valve as a cause for development of SAS has never been reported. We found that bicuspid aortic valve significantly increased the risk of recurrence of SAS in first 2 years after primary repair. Though in a small number of patients, this finding might be related to the altered haemodynamics which a bicuspid aortic valve induces in the subaortic region, leading to primary occurrence of SAS (high preoperative peak gradient) followed by its recurrence. These patients may also have an intrinsically smaller aortic valve annulus or an annular diameter that does not increase in proportion to somatic growth despite relief of SAS. A more robust study involving computational fluid dynamics in bicuspid aortic valve morphology in a larger number of patients will be required to delineate causal relations.

The risk of recurrence of SAS in this study increased proportional to the increased peak gradient across the LVOT preoperatively. Increased preoperative peak gradient across the LVOT is an independent predictor for recurrence in several studies reporting a wide range of peak gradient from >40 to 80 mmHg [12, 17, 18, 20]. In our study, the median peak LVOT gradient was 52 mmHg at the time of surgery.

There was no consistent cut-off limit of peak gradient value for referral for surgery in our study population but was assisted by clinical presentation, other echocardiography findings (including left ventricular hypertrophy, presence of aortic regurgitation) and associated cardiac lesions. Importantly, Karamlou et al. [14] suggested that subaortic resection should be delayed until the LVOT mean gradient exceeds 30 mmHg as most children with lesser gradients have a quiescent disease. To summarize, most would agree that a peak gradient of 60 mmHg or a mean gradient of 40 mmHg is an indication for surgical intervention [9], but also depends on associated echocardiographic findings.

Another risk factor for recurrence reported is aortic valve-to-membrane distance <5 mm [18]. We have observed that the subaortic membrane is a complex 3-dimensional structure rather than a simple circumferential lesion, which inconsistently adheres to any of the aortic valve cusps, making it an unreliable predictor for reintervention.

The recurrence rate for SAS in our cohort was 18% (n = 20) with a median follow-up of 13 years (IQR 7, 18) despite employing septal myectomy almost universally during the primary operation. The reported rate of recurrence has been between 13% and 31% [1, 5, 12, 15, 16]. Many patients in our study group underwent reoperation twice or even thrice for LVOTO recurrence in the form of more radical operations including Modified Konno or Ross–Konno procedures (Table 2). Consequently, although discrete and isolated in most cases at initial presentation, SAS remains a complex and progressive disease in some patients. Ten out of the 15 patients who underwent reinterventions in this cohort included procedures for aortic valvar stenosis in the form of aortic valve repair or replacement, balloon aortic valvotomy or Ross–Konno procedure signifying that SAS is part of a spectrum of multilevel LVOTO which can be progressive and challenging in some patients.

Aortic regurgitation has been associated with SAS and can predict worse outcomes [4, 12]. The preoperative peak LVOT gradient of ≥80 mmHg is an independent predictor of moderate regurgitation postoperatively [12] and its presence of regurgitation at any time-point is a significant predictor of regurgitation at later point in follow-up [27]. In our study, no covariable predicted severity of aortic regurgitation at latest follow-up.

Limitations

This study is limited by its retrospective nature and any subjective bias associated with performance and interpretation of echocardiography studies. As the included patient cohort spans over a period of almost 3 decades, there have been changes in the echocardiographic assessments, improvements in echocardiography techniques, differences in thresholds for peak gradients and differences in referral patterns for surgical correction. Also, interpretation of our results must be made while keeping in mind a small sample size and small number of reinterventions.

CONCLUSIONS

SAS remains a progressive disorder in some individuals with a significant recurrence rate associated with development of multilevel LVOTO that often demands more radical reoperations. Increased preoperative peak LVOT gradient, and presence of bicuspid aortic valve are important predictors of reoperation. Aortic regurgitation is a major consequence of SAS, but its severity usually remains low and reintervention exclusively for its management is rare.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

Funding

None.

Conflict of interest: none declared.

DATA AVAILABILITY

All relevant data are within the manuscript and its supporting information files.

Author contributions

Dushan Bandara: Conceptualization; Data curation; Formal analysis; Resources; Software; Writing—original draft; Writing—review & editing). Gananjay Gopalrao Salve: Conceptualization; Data curation; Formal analysis; Methodology; Software; Validation; Writing—original draft; Writing—review & editing. Supreet P. Marathe: Data curation; Formal analysis; Investigation; Methodology; Resources; Software; Validation; Writing—review & editing. Kim S. Betts: Data curation; Formal analysis; Investigation; Methodology; Software; Supervision; Validation; Writing—review & editing. Andrew D. Cole: Data curation; Investigation; Methodology; Project administration; Resources; Software; Validation. Julian G. Ayer: Conceptualization; Investigation; Methodology; Project administration; Supervision; Validation; Writing—review & editing. Ian A. Nicholson: Conceptualization; Methodology; Project administration; Supervision; Validation; Writing—review & editing. Yishay Orr: Conceptualization; Methodology; Project administration; Supervision; Validation; Visualization; Writing—review & editing.

Presented at the 103rd AATS Annual Meeting, Los Angeles, May 2023.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• AVR

  Aortic valve replacement

 
• CI

  Confidence interval

 
• HR

  Hazard ratio

 
• IQR

  Interquartile range

 
• LVOT

  Left ventricular outflow tract

 
• LVOTO

  Left ventricular tract obstruction

 
• SAS

  Subaortic stenosis

 
• VSD

  Ventricular septal defect