Author + information
- Received March 7, 2007
- Revision received May 30, 2007
- Accepted July 1, 2007
- Published online October 9, 2007.
- Alon Geva, AB⁎,2,
- Colin J. McMahon, MB⁎,1,2,
- Kimberlee Gauvreau, ScD⁎,
- Laila Mohammed, RDCS⁎,
- Pedro J. del Nido, MD† and
- Tal Geva, MD⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Tal Geva, Department of Cardiology, Children’s Hospital Boston, 300 Longwood Avenue, Boston, Massachusetts 02115.
Objectives This study aimed to identify independent predictors of reoperation after successful resection of discrete subaortic stenosis (DSS).
Background Recurrence of DSS has been reported to range from 0% to 55% of patients. Factors associated with recurrence have not been adequately defined.
Methods Patients were included if they had a diagnosis of DSS, normal segmental cardiac anatomy, previous resection of DSS, and at least 36 months’ follow-up. Demographic, surgical, and echocardiographic data were analyzed. Primary outcome was repeat resection of DSS in patients after successful primary resection.
Results Of 111 subjects who had successful surgical resection of DSS, 16 patients (14%) required reoperation. Median follow-up time was 8.2 years. Form of DSS and gender did not differ significantly between those with reoperation and those without. In multivariate analysis, independent predictors of reoperation that would be available before first surgery were <6 mm distance between the aortic valve (AoV) and the obstruction (hazard ratio [HR] 5.1; p = 0.013) and peak gradient by Doppler ≥60 mm Hg (HR 4.2; p = 0.016). If intraoperative variables are also considered, peeling of the membrane from the AoV or mitral valve at first surgery, <6 mm distance between the DSS and AoV, and peak gradient by Doppler ≥60 mm Hg were independent predictors of reoperation.
Conclusions Proximity of the obstructive lesion to the AoV and severe obstruction determined by preoperative echocardiography, as well as involvement of valve leaflets requiring surgical peeling, predict recurrent DSS requiring reoperation.
Discrete subaortic stenosis (DSS) encompasses a range of obstructions of the left ventricular outflow tract (LVOT) below the aortic valve (AoV) (1). The lesion is rarely clinically apparent in infancy, but the gradient across the obstruction increases progressively in approximately 50% of patients (2). Left untreated, progressive DSS can lead to left ventricular hypertrophy and eventual failure, arrhythmias, aortic valve regurgitation, and endocarditis (3–6). Although surgery is effective in the majority of patients, recurrence occurs in 0% to 55% (7–12), and reoperation is required in a fraction of those. However, risk factors for reoperation after primary resection have not been fully defined. Small cohorts, short follow-up intervals, and inconsistent definitions of recurrence limit the generalizability of earlier studies. The present study reviewed a large cohort of patients who underwent resection of DSS to determine risk factors for reoperation, with particular emphasis on preoperative prognostic indicators, based on data at medium- to long-term follow-up.
The computer database of the Cardiovascular Program at Children’s Hospital Boston, which includes information on all patients who underwent diagnostic or therapeutic procedures at our institution, was searched for all patients with a diagnosis of subvalvar aortic stenosis made between January 1984 and July 2001. Patients were included in the study if they fulfilled the following criteria: 1) a diagnosis of DSS; 2) normal segmental cardiac anatomy; 3) previous surgical resection of DSS; and 4) at least 36 months’ follow-up or reoperation for recurrent DSS within that time period. Patient records through January 2007 were reviewed for follow-up data. Patients with hypertrophic cardiomyopathy and atrioventricular (AV) canal anomalies were excluded, because the mechanisms of subaortic stenosis in those lesions differ from those in DSS. Demographic data including gender, age at diagnosis, and age at first surgical resection were abstracted from the medical records. The Children’s Hospital Boston Committee on Clinical Investigations approved review of the medical records.
Throughout the study period, the most common surgical approach to DSS resection at our institution was an endarterectomy type of resection using a blunt stripping technique to remove fibroelastic tissue. Sharp excision was used to perform limited myomectomy in cases judged to be appropriate by the surgeon. The following intraoperative details were recorded from the surgical reports: performance of myomectomy, peeling or scraping of adherent subaortic fibrous tissue off the aortic and/or mitral valve (MV) leaflets, and concomitant repair of other cardiac lesions (e.g., ventricular septal defect closure, MV repair).
The initial, preoperative, postoperative, and most recent echocardiograms were reviewed, and selected still frames were used for analysis. Measurements were obtained in triplicate, and means were used for statistical analysis. Height and weight were recorded and body surface area was calculated using the Haycock formula (13) at each echocardiogram.
The DSS morphology was categorized as membrane, fibromuscular ridge (FMR), or FMR with associated membrane. Abnormal AoV morphology (categorized as thickened leaflets, unicuspid or bicuspid valve, or doming during systole) was noted. Extension of the membrane onto the aortic valve leaflets was noted when present. Any involvement of the mitral valve or submitral apparatus with attachments to the LVOT was recorded. Distance of the lesion from the hinge point of the right coronary cusp of the AoV was measured in systole and diastole from the parasternal long-axis view as described by Kleinert and Geva (14). Other measurements included: length of the membrane or FMR, measured from the parasternal long-axis view as the distance from the septal insertion to the free edge of the obstructive lesion; AoV annulus diameter zscore; aortoseptal angle, measured as described by Kleinert and Geva (14) and Fowles et al. (15); and distance between the aortic and mitral valves in diastole. Degree of override of the aorta was graded qualitatively as normal when less than one-third of the annulus was intercepted by an extrapolation of the ventricular septum long axis, as mild when one-third to two-thirds of the annulus was intercepted, and as marked when more than two-thirds was intercepted. Associated cardiac lesions, if any, were recorded. The maximum instantaneous gradient across the LVOT, calculated using the modified Bernoulli equation, was recorded from spectral Doppler tracings. Aortic regurgitation (AR) was graded qualitatively as none, trivial, mild, moderate, or severe as described by Tani et al. (16).
The primary outcome was repeat surgical resection of DSS in patients after successful primary resection, defined as early (<1 month) postoperative transthoracic Doppler gradient <40 mm Hg. A secondary outcome was DSS recurrence, defined as peak LVOT gradient ≥40 mm Hg at any time after the first postoperative month.
Patient and clinical characteristics were compared between subjects who underwent repeat surgical resection of DSS after an initial successful operation and those who did not using Fisher exact test for categoric variables and the Wilcoxon rank sum test for continuous variables. Similarly, characteristics were compared for patients who did and did not develop recurrence. Freedom from reoperation was estimated using the Kaplan-Meier method; patients who did not require reoperation were censored at the time of last follow-up. Univariate analyses examining relationships between time to reoperation and categoric risk factors were evaluated using the log rank test. Appropriate cut points for continuous risk factors were explored descriptively and evaluated in the same way. Additional univariate and multivariate analyses for predictors of time to reoperation were conducted using the Cox proportional hazards model. Analyses exploring characteristics associated with DSS recurrence were performed using logistic regression analysis, controlling for length of follow-up. To determine whether absolute or BSA-adjusted DSS to AoV distance was a better predictor of reoperation, logistic regression models adjusting for length of follow-up were fitted and areas under the receiver-operating characteristic curves compared. Interquartile ranges were calculated for median peak pressure gradients at different time points. A commercially available statistical software package was used for data analysis (Stata version 9.0, StataCorp, College Station, Texas).
Of 1,187 patients who had some form of subaortic stenosis, 219 had DSS. Of these patients, 112 (51%) underwent surgical resection during the study period. Initial surgery included resection of the obstructive subaortic lesion in all patients. In 58 patients (52%), associated septal myomectomy was performed, and septal augmentation (modified Konno procedure) was performed in 4 (4%). The surgeon removed fibrous tissue extending from the DSS to the AoV or MV in 12 (11%) and 7 (6%) of the patients, respectively (Table 1).Additional surgical procedures for associated congenital cardiac anomalies (e.g., ventricular septal defect closure, coarctation repair) were performed at the time of DSS resection in 23 patients (21%). One patient subsequently had an early postoperative gradient ≥40 mm Hg, which we defined as a failed primary resection.
Indications for DSS resection varied but typically involved agreement between the patient’s cardiologist and cardiac surgeon that the severity of the LVOT obstruction or degree of AR warranted intervention. Highest peak gradient before surgery did not change significantly during the 17-year study period (Fig. 1).In 34 patients (31%), the primary indication was peak LVOT Doppler gradient >35 mm Hg alone, and in 6 patients (5%) at least mild AR prompted intervention. Forty-four patients (40%) had both high LVOT Doppler gradient and at least mild AR. In addition, some patients underwent DSS resection in the course of surgery for other associated cardiac anomalies. Of these patients, in 16, the peak LVOT Doppler gradient was >35 mm Hg, and in 7 it was ≤35 mm Hg. Four patients underwent DSS resection for other reasons. Left ventricular hypertrophy (left ventricular posterior wall thickness zscore >2) was present in 40% of patients.
Sixteen patients (14%) required reoperation. Median time to reoperation was 6.9 years (range 1.7 to 11.2 years). Indications for reoperation included recurrent obstruction with or without worsening AR and with or without left ventricular hypertrophy. The highest peak gradient before reoperation did not vary significantly during the study period. Of the 16 patients who had reoperations, one had an implantable defibrillator after a third resection and 2 had an iatrogenic ventricular septal defect.
An additional 11 patients had postoperative gradients during follow-up ≥40 mm Hg but did not undergo further resection during the study period. Demographic and clinical characteristics for the cohort as well as for the groups of patients with and without repeat DSS resection are summarized in Table 1. None of the patients in this cohort died owing to DSS or its treatment. Three patients (3%) had complete AV block and required placement of an implantable pacemaker.
Predictors of reoperation
Form of DSS and gender did not differ significantly between those who did and did not require reoperation. In univariate analysis, younger age at first surgery, smaller AoV annulus diameter zscore, shorter distance between the DSS obstruction and the AoV, and higher peak gradient across the obstruction on preoperative echocardiogram were significantly associated with subsequent reoperation. Additionally, peeling of the obstructive fibrous tissue from the AoV or MV and myomectomy during the initial surgery were associated with future reoperation. Use of septal augmentation (Konno procedure) did not differ significantly between the 2 groups. The only postoperative echocardiographic variable associated with reoperation was higher peak gradient across the LVOT. Diagnosis of Shone syndrome (defined as ≥3 left heart obstructive lesions) was significantly associated with future reoperation for recurrent DSS. No other cardiac lesions, including aortic stenosis, abnormal AoV morphology, thickened AoV leaflets, doming AoV, or MV involvement, were associated with the primary outcome. Table 2shows variables predictive of reoperation by univariate analysis.
In multivariate analysis using the Cox proportional hazards model and considering only variables that would be available before first surgery (Table 3),independent predictors of earlier time to reoperation were distance between the obstructive lesion and the AoV <6 mm in systole (hazard ratio [HR] 5.1, 95% confidence interval [CI] 1.4 to 18.3; p = 0.013) and peak systolic gradient by Doppler ≥60 mm Hg (HR 4.2, 95% CI 1.3 to 13.5; p = 0.016). If both pre- and intraoperative variables are considered for the model, peeling of the membrane from the AoV or MV at first surgery (HR 4.2, 95% CI 1.4 to 12.6; p = 0.012), <6 mm distance between the DSS and AoV in systole (HR 4.1, 95% CI 1.1, 15.0; p = 0.032), and peak systolic gradient by Doppler ≥60 mm Hg (HR 3.8, 95% CI 1.2 to 12.2; p = 0.026) were independent predictors of shorter time to reoperation.
Figure 2shows a Kaplan-Meier curve illustrating overall freedom from reoperation for the cohort; Figures 3Aand 3B show how freedom from reoperation differed for those patients with shorter distance between the DSS and AoV and for those with higher peak gradients on preoperative echocardiogram, respectively.
Predictors of recurrence
In 27 patients with a successful primary resection, a maximum instantaneous gradient across the LVOT ≥40 mm Hg was noted on follow-up echocardiogram. Younger age at diagnosis and at first surgery, associated Shone syndrome, peak preoperative gradient ≥70 mm Hg, smaller AoV annulus diameter zscore, shorter distance between the DSS obstruction and the AoV, and peeling of the membrane from the AoV or MV during surgery were significantly associated with recurrence by univariate analysis. Multivariate analysis found predictors of recurrence that were similar to predictors of reoperation (Table 4).However, associated Shone syndrome was a slightly stronger predictor (HR 3.6, 95% CI 1.0 to 12.5; p = 0.048) and thus replaced higher peak preoperative gradient in the model.
Comparison between the 16 patients with reoperation and the 11 patients with recurrence but no reoperation showed no significant differences between the groups in demographic and most clinical characteristics. The only statistically significant exception was a shorter DSS-AoV distance in those who had reoperation (median 5.5 vs. 6.4 mm in systole; p = 0.032).
Comparison by age at first surgery
To test the hypothesis that patients with younger age at first surgery had more aggressive underlying pathology, we compared those who were younger than 4 years old at first surgery (n = 32) to those who were at least 4 years old (n = 79). Younger patients had Shone syndrome more often (38% vs. 8%; p < 0.001), had shorter distances between the DSS and the AoV (5.8 vs. 7.8 mm in systole; p = 0.005), and tended to have the membrane peeled from the AoV or MV during surgery more often (28% vs. 13%; p = 0.09). As expected, DSS was diagnosed earlier in patients who were operated at age <4 years (1.1 vs. 5.4 years; p < 0.001). Although they had significantly lower gradients across the LVOT at baseline (31 vs. 45 mm Hg; p = 0.057), patients who were operated on at a younger age had peak gradients during preoperative echocardiography as high as their older counterparts (49 vs. 50 mm Hg; p = NS) (Fig. 4).
In this study, we have analyzed data on a large cohort of patients with relatively long postoperative follow-up of up to 20.7 years (median 8.2 years) to determine independent predictors of reoperation for recurrent discrete subaortic stenosis after successful primary resection. The present study used reoperation rather than recurrence as a concrete clinically relevant outcome. Earlier studies, which report recurrence rates ranging from 0% to 55%, are difficult to compare, because they use different criteria for recurrence of DSS or, often, report no specific criteria at all (7,9–12). Nonetheless, when looking at recurrence as a secondary outcome, we identified independent predictors similar to those identified for reoperation.
Although some previous studies have suggested that performing myomectomy during first surgery reduces the incidence of recurrence (9,17–19), other authors have questioned this finding (7,20). In the present study, patients who underwent myomectomy showed a trend toward significantly greater risk of reoperation. These differences may be due to different outcome definitions used in previous studies or the fact that surgeons’ perception of need for myomectomy corresponds to more aggressive underlying disease that was accounted for by other variables in our multivariate model. Importantly, in a study by Parry et al. (9), which reported no recurrence using an aggressive surgical resection with peeling of the membrane from the AoV and extensive myomectomy, the rate of AV block was 14% versus the typical 1% to 5% AV block rate reported in most studies (7,10–12), including the present one (3%). The experience of Parry et al. (9) thus highlights the tradeoff between the risk of AV block and a potentially lower recurrence rate associated with extensive myomectomy.
Higher peak gradient across the LVOT at preoperative echocardiogram was an independent predictor of need for reoperation. Testing various cut-off points, we found that a peak gradient ≥60 mm Hg is most predictive of need for reoperation. Higher peak pre- and postoperative gradients by Doppler are the most common risk factors reported for recurrence (7,9,20,21), likely indicating a more aggressive DSS pathology.
Discrete subaortic stenosis is a progressive disease that is often not diagnosed in infancy. Nonetheless, patients with earlier age at diagnosis and earlier age at first surgery were more likely to need reoperation. Although this might suggest that age at first surgery is a predictor of need for reoperation, multivariate logistic regression analysis eliminates it as an independent variable. Our analysis suggests that this occurs because younger age at first surgery is a surrogate for a more aggressive disease process. Patients with younger age at first surgery had significantly faster progression of peak gradient across the LVOT (Fig. 4), DSS significantly closer to the AoV, and significantly higher incidence of Shone syndrome. Notably, these variables were significant independent predictors of reoperation and recurrence in the overall cohort. These findings suggest that patients with more rapidly progressing disease and more likely need for reoperation might have a different underlying pathophysiology than those whose clinical course progresses more insidiously. Laboratory studies might be used to address possible biologic differences in the obstructive subaortic lesion.
Several limitations of this study merit attention. As a retrospective case-control study, its predictions are based on any systematic bias in the study cohort. Also, criteria for initial or subsequent operation were not standardized. Although we conducted multivariate regression analysis that controls for identifiable confounding variables, our findings need to be validated by prospective studies testing this study’s predictive power. In addition, many of our observations regarding differences in disease manifestation in patients with younger age at first surgery show a trend toward statistical significance. Because only 32 patients were <4 years old, our cohort may not have statistical power to detect subtle differences in pathophysiology leading to earlier first surgery. Future studies should examine the underlying pathophysiology that leads to more aggressive disease and need for reoperation.
The findings of this study suggest that patients with higher peak gradient at preoperative echocardiogram and those with shorter distance between the DSS and the AoV require close follow-up after their first surgery. Those who require peeling of the membrane from the AoV or the MV during surgery are also at greater risk of needing subsequent reoperation, and their management should be adjusted accordingly. Given that myomectomy did not reduce the risk of reoperation in this cohort, further refinements of surgical techniques are needed to decrease the likelihood of DSS recurrence safely.
- Abbreviations and Acronyms
- aortic valve
- aortic regurgitation
- discrete subaortic stenosis
- fibromuscular ridge
- left ventricular outflow tract
- mitral valve
- Received March 7, 2007.
- Revision received May 30, 2007.
- Accepted July 1, 2007.
- American College of Cardiology Foundation
- Katz N.M.,
- Buckley M.J.,
- Liberthson R.R.
- Somerville J.,
- Stone S.,
- Ross D.
- Brauner R.,
- Laks H.,
- Drinkwater D.C. Jr..,
- Shvarts O.,
- Eghbali K.,
- Galindo A.
- Parry A.J.,
- Kovalchin J.P.,
- Suda K.,
- et al.
- Rohlicek C.V.,
- del Pino S.F.,
- Hosking M.,
- Miro J.,
- Cote J.M.,
- Finley J.
- de Vries A.G.,
- Hess J.,
- Witsenburg M.,
- Frohn-Mulder I.M.,
- Bogers J.J.,
- Bos E.
- Kleinert S.,
- Geva T.
- van Son J.A.,
- Schaff H.V.,
- Danielson G.K.,
- Hagler D.J.,
- Puga F.J.
- Coleman D.M.,
- Smallhorn J.F.,
- McCrindle B.W.,
- Williams W.G.,
- Freedom R.M.