Article Text

Download PDFPDF

Original article
Long-term outcomes and risk factors for aortic regurgitation after discrete subvalvular aortic stenosis resection in children
  1. Sarah S Pickard1,2,
  2. Alon Geva3,4,
  3. Kimberlee Gauvreau1,2,
  4. Pedro J del Nido5,6,
  5. Tal Geva1,2
  1. 1Departments of Cardiology, Boston Children's Hospital, Boston, Massachusetts, USA
  2. 2Departments of Pediatrics, Harvard Medical School, Boston, Massachusetts, USA
  3. 3Critical Care Medicine, Boston Children's Hospital, Boston, Massachusetts, USA
  4. 4Department of Anesthesia, Harvard Medical School, Boston, Massachusetts, USA
  5. 5Cardiovascular Surgery, Boston Children's Hospital, Boston, Massachusetts, USA
  6. 6Department of Surgery, Harvard Medical School, Boston, Massachusetts, USA
  1. Correspondence to Dr Tal Geva, Department of Cardiology, Boston Children's Hospital, 300 Longwood Avenue, Boston, MA 02115, USA; tal.geva{at}


Objectives To characterise long-term outcomes after discrete subaortic stenosis (DSS) resection and to identify risk factors for reoperation and aortic regurgitation (AR) requiring repair or replacement.

Methods All patients who underwent DSS resection between 1984 and 2009 at our institution with at least 36 months’ follow-up were included. Demographic, surgical and echocardiographic data were reviewed. Outcomes were reoperation for recurrent DSS, surgery for AR, death and morbidities, including heart transplant, endocarditis and complete heart block.

Results Median length of postoperative follow-up was 10.9 years (3–27.2 years). Reoperation occurred in 32 patients (21%) and plateaued 10 years after initial resection. Survival at 10 years and 20 years was 98.6% and 86.3%, respectively. Aortic valve (AoV) repair or replacement for predominant AR occurred in 31 patients (20%) during or after DSS resection. By multivariable analysis, prior aortic stenosis (AS) intervention (HR 22.4, p<0.001) was strongly associated with AoV repair or replacement. Risk factors for reoperation by multivariable analysis included younger age at resection (HR 1.24, p=0.003), preoperative gradient ≥60 mm Hg (HR 2.23, p=0.04), peeling of membrane off AoV or mitral valve (HR 2.52, p=0.01), distance of membrane to AoV <7.0 mm (HR 4.03, p=0.03) and AS (HR 2.58, p=0.01).

Conclusions In this cohort, the incidence of reoperations after initial DSS resection plateaued after 10 years. Despite a significant rate of reoperation, overall survival was good. Concomitant congenital AS and its associated interventions significantly increased the risk of AR requiring surgical intervention.

Statistics from


Discrete subaortic stenosis (DSS) is an often-progressive obstruction of the LV outflow tract (LVOT) that, without intervention, may result in LV hypertrophy and dysfunction, aortic regurgitation (AR), endocarditis, arrhythmias and death.1–4 Despite successful surgical resection, 8–31% of patients require reoperation due to recurrent obstruction.4–14 We and other investigators have previously reported midterm results of DSS resection and identified risk factors for reoperation.8 However, the average length of postoperative follow-up in these series was relatively short.6 ,8 ,10–13 ,15–17 Given the paucity of long-term follow-up in the paediatric population, it is unknown whether the rate of DSS recurrence requiring reoperation persists or plateaus beyond the first postoperative decade.

Another long-term concern in patients with DSS is deterioration of aortic valve (AoV) function manifesting predominantly as regurgitation. Furthermore, detection of AR by echocardiography often triggers referral for DSS resection. However, little information exists regarding factors associated with haemodynamically significant AR in these patients. The high prevalence of AR (40–80%) has been attributed to the jet-effect on the aortic leaflets created by the obstructive lesion, as noted by studies demonstrating an association between higher preoperative gradient and increased risk of AR progression.2 ,4 ,11 ,13 ,15 ,18 ,19 However, few studies have examined the effects of concomitant congenital anomalies of the AoV on AR in this group of patients. To address these gaps in knowledge, this study was undertaken to characterise long-term outcomes after DSS resection and to identify risk factors for late reoperation and AR requiring repair or replacement.



Patients with DSS who had undergone surgical resection between January 1984 and May 2009 were included if they fulfilled the following criteria: (1) a diagnosis of DSS; (2) normal segmental cardiac anatomy; and (3) at least 36 months follow-up. ‘Isolated’ DSS was defined as subaortic stenosis without additional anatomical lesion with the exception of a spontaneously closed ventricular septal defect (VSD). Patients with hypertrophic cardiomyopathy and atrioventricular canal anomalies were excluded. Gender, age at diagnosis, age at surgical resection and complications (date of death, cause of death, transplantation, endocarditis, complete heart block (CHB) requiring pacemaker) were abstracted from the medical records. The Boston Children's Hospital Committee on Clinical Investigations approved review of the medical records and waived the requirement for informed consent.

Surgical and catheter data

Indication for operation was determined by consensus opinion of the patient's cardiologist and surgeon, incorporating age, degree and rate of progression of obstruction, AR, associated lesions, and comorbidities. The general operative approach remained consistent throughout the study period with endarterectomy-type resection using a blunt stripping technique to remove fibroelastic tissue.8 The following intraoperative details were recorded: performance of myomectomy, peeling or scraping of adherent subaortic fibrous tissue off the aortic and/or mitral valve leaflets, and additional surgical procedures (aortic commisurotomy, valvotomy, valvuloplasty, replacement, Ross-Konno, or patch augmentation of the LVOT (Konno)). Surgical procedures on the AoV for the purpose of relief of valvular stenosis were recorded. When available, catheter reports were reviewed for performance of balloon aortic valvuloplasty.

Echocardiographic data

The preoperative, postoperative and most recent echocardiograms were reviewed. DSS morphology was categorised as membranous, fibromuscular ridge (FMR) or FMR with associated membrane, as previously described.8 AoV morphology was noted. The highest preoperative maximum instantaneous gradient across the LVOT was recorded from spectral Doppler tracings. AR was graded qualitatively as none, trivial, mild, moderate or severe as described by Tani et al.20 Distance of the lesion from the hinge point of the right coronary cusp of the AoV was measured in systole from the parasternal long-axis view.21


The outcomes of interest included: (1) repeat surgical resection of DSS after successful initial resection (defined as early (<1 month) postoperative maximum instantaneous Doppler gradient ≤40 mm Hg); (2) recurrence (gradient ≥40 mm Hg); (3) surgical procedures for AR with or after DSS resection, including valve repair, pulmonary autograft (Ross procedure) or replacement; and (4) adverse postoperative outcomes, including (A) early death (defined as death prior to hospital discharge), (B) late death, (C) cardiac transplantation, (D) CHB requiring pacemaker and (E) endocarditis.

Statistical analysis

Patient and clinical characteristics were summarised for subjects who underwent repeat surgical resection and those who did not. Comparisons of time to repeat surgical resection were performed using the log-rank test for categorical variables and the Cox proportional hazards model for continuous variables. Freedom from outcome events following initial resection was estimated using the Kaplan–Meier method for reoperation, death, and AoV repair or replacement. Patients who did not experience the outcome of interest were censored at the time of last follow-up or transplantation. Cox models were used to examine relationships between patients and clinical characteristics and times to reoperation and AoV repair or replacement. Factors significant at the 0.10 level of significance in unadjusted analyses were considered for inclusion in multivariable models. Forward stepwise selection was used, and a p value ≤0.05 was required for retention in the final models. Aortic stenosis (AS) intervention was treated as a time-varying covariate. Sensitivity analyses treating death as a competing risk rather than a censoring event were performed using the method of Fine and Gray.22 Receiver-operator characteristic curves were used to identify the best cut-point for relating distance of DSS to AoV to outcomes. Commercially available software packages were used for analysis (SPSS V.19.0, IBM Corp., Chicago, Illinois, USA; Stata V.13.1, StataCorp LP, College Station, Texas, USA).



Table 1 summarises the demographic and clinical characteristics of the 155 study patients. Median age at diagnosis was 2.5 years (range, birth to 34.8 years) and median age at initial surgical resection was 5.2 years (range, 0.1–35 years). Duration of postoperative follow-up ranged from 3.0 years to 27.2 years (median 10.9 years). The majority of patients (70%) had a membranous (fibrous) form of DSS with the remaining having either a thick FMR or a FMR with a fibrous membrane at the edge of the lesion. The median preoperative maximum instantaneous gradient across the LVOT measured 55 mm Hg (range, 15–150 mm Hg). Associated cardiac anomalies were common and only 55 patients (35.5%) had ‘isolated’ DSS. Valvular AS was the most frequently associated anomaly, occurring in 51 patients (33%), often in association with bicommissural or unicommissural valve morphology.

Table 1

Demographic, anatomical and surgical characteristics

Initial DSS resection

The surgical technique during initial DSS resection included endarterectomy-type resection with blunt stripping of the obstructive tissue in all patients. Additional myomectomy was performed in 78 patients (50%) and peeling of fibroelastic tissue from the ventricular aspect of the aortic, mitral or both valves was done in 15%, 23% and 3% of patients, respectively. Median early postoperative maximum instantaneous Doppler gradient across the LVOT decreased to 15 mm Hg.


Of 155 patients, 32 (20.6%) underwent reoperation for recurrent DSS at a median of 6.3 years after initial resection (table 1). Univariable predictors of reoperation included younger age at diagnosis and initial resection, higher preoperative and postoperative LVOT gradients, distance of the obstruction to AoV <7.0 mm, and presence of AS, Shone syndrome or mitral stenosis (table 1). Length of postoperative follow-up was similar in those who did and did not undergo reoperation (p=0.11). By multivariable analysis (table 2), younger age at DSS resection, preoperative gradient ≥60 mm Hg, peeling of the membrane, distance of the obstruction to AoV and AS predicted time to reoperation. Four patients (2.6%) underwent a third resection. There was no difference in rate of reoperation between those with and without ‘isolated’ DSS. Freedom from reoperation at 5 years, 10 years and 20 years was 91.3% (95% CI 85.4% to 94.9%), 76.7% (95% CI 68.0% to 83.3%) and 71.7% (95% CI 61.2% to 79.8%), respectively (figure 1A). Reoperations plateaued approximately 10 years after initial resection; over 90% of reoperations occurred in the first 10 years.

Table 2

Multivariable analysis of predictors of time to DSS reoperation

Figure 1

Kaplan–Meier survival curves for freedom from reoperation (A) and cumulative survival (B). Dashed lines indicate 95% CIs.


Recurrence of DSS with a peak Doppler gradient ≥40 mm Hg occurred in 38 patients (24.5%). Of those, 32 underwent reoperation.


Table 3 summarises morbidity and mortality. Two patients were transplanted. One underwent a Ross operation complicated by dehiscence of the right coronary anastomosis with resultant ventricular dysfunction and transplantation 1 year later. The other patient had pulmonary hypertension and RV dysfunction after repeat DSS resection. Transplantation occurred 27 days after resection and the patient died 3 days later of RV failure. CHB resulting in pacemaker placement occurred in seven patients (4.5%) following initial or repeat DSS resection: three were concurrent with primary DSS resection, two were concurrent with repeat resection with a Ross-Konno procedure, one had prior VSD repair and one had concurrent VSD repair. There were no episodes of postoperative heart block requiring pacemaker since 2000. Endocarditis occurred in three patients (1.9%).

Table 3

Morbidity and mortality after DSS resection in 155 patients


There was no early mortality after initial DSS resection. There were seven all-cause late deaths. Two patients died out of hospital following cardiac arrest (confirmed ventricular fibrillation in one), one died of complications of restrictive lung disease, one died soon after transplantation for severe pulmonary hypertension and RV failure, and three had unknown causes of death. Survival rates did not differ between those with and without ‘isolated’ DSS. Survival from initial DSS resection at 1 year, 5 years, 10 years and 20 years was 100%, 99.3% (95% CI 95.2% to 99.9%), 98.6% (95% CI 94.4% to 99.6%) and 86.1% (95% CI 66.5% to 94.6%), respectively (figure 1B).

Aortic valve stenosis

Associated AS was diagnosed in 51 patients (33%). Of these, 33 (21% of the entire cohort and 65% of patients with AS) underwent interventions to relieve the AS. Table 4 summarises percutaneous and surgical interventions for AS prior to, with and after DSS resection. Only four patients who had not required AS intervention prior to DSS resection went on to have an AS intervention after the first DSS resection.

Table 4

Aortic valve interventions

Aortic regurgitation

Table 4 summarises the surgical procedures for AR. During initial DSS resection, 19 patients (12.3%) required surgical intervention for predominant AR. Following DSS resection, 15 patients (9.7%) developed sufficient regurgitation to require surgical repair or replacement. Only one patient in the ‘isolated’ DSS group required intervention on the AoV for significant regurgitation. At last follow-up prior to AoV intervention for significant AR, 18.7% had moderate or greater AR.

Univariable predictors of AR requiring repair or replacement are summarised in table 5. Notable factors associated with the outcome included valvular AS, AS interventions, bicuspid/unicuspid AoV, mitral stenosis, Shone syndrome and peak preoperative gradient ≥60 mm Hg. Multivariable analysis identified AS intervention (HR 22.4, p<0.001) as an independent predictor of AoV repair or replacement during or after DSS resection (table 5). Figure 2 demonstrates freedom from AoV repair or replacement stratified by presence of AS.

Table 5

Predictors of aortic valve repair or replacement with or after DSS resection

Figure 2

Kaplan–Meier curves comparing lifetime freedom from AoV surgery for aortic regurgitation (AR) with and without aortic stenosis (AS). Patients with concurrent AS were significantly more likely to require AoV surgery at any point during their lifetime.


In this large single-centre cohort of patients who underwent DSS resection and were followed for up to 27 years after surgery, we found that (1) long-term survival was good, with 86.3% alive at 20 years; (2) recurrence of DSS requiring reoperation occurred in 21% of patients, with most reoperations occurring within the first 10 years after initial resection; and (3) concomitant valvular AS was common and the procedures performed for its treatment are strongly associated with AR of sufficient severity to warrant an operation.

Survival and morbidity

We demonstrate an overall excellent prognosis with no early postoperative deaths after initial resection. Our survival estimate is consistent with prior analyses of the paediatric population.23 Van der Linde et al4 reported a higher 20-year survival of 97% in an adult cohort, which is likely secondary to lower anatomical complexity (only 3.2% of their patients had Shone complex, compared with 14.8% in ours). CHB requiring pacemaker was relatively uncommon. Interestingly, there were no cases after 2000 while the rate of concurrent myomectomy since then has not changed. In contrast to prior studies, there was no association between concurrent myomectomy and need for pacemaker placement.4 ,23 Therefore, the lower rate of CHB compared with prior studies and apparent reduction in this complication over time is likely due to changes in surgical technique that avoid the conduction axis rather than abandonment of an extended myocardial resection.5 ,7 ,23


While the overall rate of reoperation for DSS has been reported by several groups,1 ,3 ,6 ,8 ,10–13 ,16 ,17 ,24–27 lack of long-term follow-up has limited the examination of the time course of reoperation after initial resection. The rate of reoperation in our cohort is consistent with prior reports and slightly higher than our 2007 analysis, as might be expected with a longer follow-up period.4 ,10–12 ,14 ,23 ,24 ,27 Jou et al23 reported a lower rate of reoperation (15.7%) in a cohort followed for a median follow-up of 4.6 years who underwent aggressive initial resection, as evidenced by a high rate of postoperative CHB (11.4%). In contrast, Valeske et al reported a higher reoperation rate of 31% during a median follow-up of 7.5 years, possibly due to the inclusion of a significant proportion of patients (48%) with long-segment, tunnel-like obstruction.10 The finding that reoperation is associated with younger age at diagnosis and initial surgery, higher preoperative gradient, peeling of the membrane, shorter distance of the obstruction to the AoV and AS is consistent with those of other groups and may suggest a more aggressive disease phenotype.4 ,6 ,8 ,17 ,24 ,27 While our 2007 analysis identified several of the above predictors for reoperation,8 the current study with its longer duration of follow-up reveals younger age and AS to be additional important predictors. A recent systematic review by Etnel et al questions the prognostic relevance of distance of the obstruction to AoV as Paul et al28 found a longer distance to be predictive of mitral regurgitation. We did not specifically investigate this relationship; however, we have repeatedly found increased risk of reoperation with shorter distance and consider it to be clinically important.8 ,15 ,29

With a median follow-up of 10.9 years, the present study demonstrates that the rate of reoperation in the paediatric population plateaus after 10 years. In contrast, van der Linde et al, who studied a large adult cohort, found a similar overall rate of reoperation (25.6%) but at a steady rate of 1.8% per patient-year without plateau.4 We speculate that the discrepancy in the timing of reoperation can be attributed to a more aggressive disease progression in young children, and thus reoperations occurred earlier. It is also conceivable that in adults the disease process in the LVOT continues to be active, albeit at a slower rate as compared with young children. Understanding age-related differences in the time course and rates of recurrent DSS and reoperation is useful for fine-tuning postoperative follow-up and informs discussion of prognosis. Furthermore, while the length of follow-up in this study is one of the longest among paediatric cohorts of DSS, it remains relatively short relative to the life expectancy of these patients. Hence, it is conceivable that after a period of low incidence of recurrence requiring reoperation 10 years after the first operation, there will be resurgence in the rate of these outcomes.

Aortic regurgitation

Several investigators have reported the frequency of AR in patients with DSS and evaluated potential aetiological factors. Comparison of rates of AR among published DSS cohorts is confounded by variable inclusion of patients with valvular AS. While Uysal et al27 reported moderate AR in only 8.7% of patients, they excluded patients with valvular AS. In contrast, Ruzmetov et al12 reported 13% of their cohort had valvular AS and 18% had moderate AR at postoperative follow-up. Their findings are consistent with our observation of significant AR requiring surgery in 20% of patients.

AR in patients with DSS has often been attributed to a jet effect and turbulence created by the obstructive subaortic tissue that progressively damages the valve leaflets, ultimately resulting in incompetence.18 This hypothesis has led many clinicians to recommend surgical resection of DSS to circumvent the development or worsening of AR. Prior studies have recommended early intervention based on the finding that a higher preoperative LVOT gradient is associated with postoperative progression of AR.4 ,6 ,11 ,14 ,30 However, peak preoperative gradient ≥60 mm Hg did not reach statistical significance in our model. There has been limited prior analysis of the roles of concomitant congenital valvular AS and its management in the setting of DSS.15 We found that intervention for valvular AS was strongly and independently associated with clinically important AR. This is consistent with a study by McMahon et al15 that found prior intervention for AS to be strongly associated with AR in patients with DSS. While there are sufficient data to suggest the degree of subaortic obstruction likely impacts the development or evolution of significant AR, the present study provides evidence that AS and its associated interventions are the major contributors to aortic valve dysfunction in this population.


Because of the retrospective nature of this study, we cannot exclude residual confounding by unmeasured covariates. Criteria for surgical intervention for DSS, AS or AR lacked standardisation. Furthermore, although the size of our cohort is among the largest with this disease studied to date, limitations in statistical power may hinder our ability to identify additional independent contributors to reoperation and AR. We cannot rule out that longer follow-up would identify DSS recurrence more than a decade after initial resection. Therefore, larger, longer-term studies are warranted to address the possibility of late resurgence of DSS requiring reoperation, late morbidity, or other potential predictors of morbidities, including AR.


The findings of this study suggest reoperation after DSS resection plateaus at a low level 10 years after initial resection; patients with more aggressive disease, as evidenced by higher preoperative LVOT gradients, younger age, or associated AS, were more likely to require reoperation. Despite a significant number of patients requiring reoperation, overall survival is good. Concomitant congenital valvular AS and its associated interventions are major contributors to AR requiring surgical intervention after DSS resection and efforts focused at refining treatment for AS should be made alongside surgery for DSS.

Key messages

What is already known on this subject?

  • Despite successful initial resection of discrete subaortic stenosis, 8–31% of patients require reoperation for recurrent obstruction. Progressive aortic regurgitation (AR) is an additional cause of reintervention. Long-term outcomes in children remain incompletely defined with ongoing concerns for late reoperation and AR requiring valve repair or replacement.

What might this study add?

  • This study characterises long-term outcomes (up to 27 years, median 11 years) and identifies risk factors for significant AR in a large cohort of paediatric patients (n=155). Reoperations occurred in 21%, plateauing after 10 years. Overall survival at 10 years and 20 years was 98.6% and 86.3%, respectively. Prior intervention for associated valvular aortic stenosis (AS), rather than severity of subaortic stenosis, independently predicts AR.

How might this impact on clinical practice?

  • Understanding age-related differences in time course of reoperation informs follow-up and prognostication. The finding of the important role of AS challenges the conventional wisdom that early resection of subaortic stenosis must be undertaken to prevent AR.


The authors thank Emily Harris for her expert assistance with the figures.



  • Contributors SSP: study design, data acquisition, analysis and interpretation, and writing and final approval of the manuscript. AG: data analysis and interpretation, editing and final approval of the manuscript. KG: study design, data analysis and final approval of the manuscript. PJdN: responsible for the surgical aspect of the project, interpretation of data and final approval of the manuscript. TG: responsible for all aspects of the project.

  • Funding Supported in part by the Higgins Family Non-invasive Cardiac Imaging Research Fund.

  • Competing interests None declared.

  • Ethics approval IRB approval.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.