The journey from repair to replacement in pediatric aortic valve stenosis: a 10-year single-center experience

Abstract

Objective

The optimal reoperation strategy and long-term outcomes of pediatric patients with congenital aortic stenosis (AS) have not been well elucidated. This study aimed to evaluate the reintervention outcomes and long-term prognosis in patients with isolated AS following their initial aortic valve (AoV) repair.

Methods

A retrospective analysis was conducted on the clinical data of all patients with isolated AS who underwent initial AoV repair between 2013 and 2024. The primary outcome was the rate of freedom from reoperation after the initial procedure and secondary surgeries.

Results

A total of 203 patients who underwent initial AoV repair were included. The median age at initial surgery was 2.4 (0.6, 4.7) years. The 30-day mortality rate was 0.5% (1/203). The 10-year freedom from AoV reoperation rate was 50.9% (95% CI: 36.2%-65.6%; n = 48/203). The 10-year freedom from AoV replacement rate was 62.7% (95% CI: 46.0%-79.4%; n = 28/203). Among the 48 patients who underwent reoperation, 20 underwent AoV repair and 28 underwent AoV replacement. Patients in the AoV replacement group were older (9.4 ± 4.0 years vs. 6.7 ± 3.6 years, P = 0.022) and had higher body weight (32.2 ± 14.3 kg vs. 22.8 ± 12.1 kg, P = 0.021). The 10-year freedom from AoV reoperation rate after the secondary procedure was 87.9% (95% CI: 73.4%-100.0%; n = 3/45). The 10-year freedom from moderate or greater AS/ aortic regurgitation was 50.2% (95% CI: 30.8%-69.6%; n = 15/45). Compared with secondary AoV repair, AoV replacement was associated with superior long-term outcomes (P = 0.001).

Conclusion

Favorable outcomes are achieved in patients with AS after initial AoV repair; however, nearly half of the patients may require reoperation within 10 years. For reoperations, AoV replacement is more likely to provide desirable long-term quality of life.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12887-025-06494-8.

Introduction

Congenital aortic stenosis (AS) in infants is a lifelong condition, and most patients may face the risk of requiring multiple interventions [1]. For surgeons, the primary consideration is a treatment approach that effectively relieves stenosis and improves long-term survival outcomes. Balloon valvuloplasty (BVPL) was previously regarded as an effective treatment option for alleviating stenosis; however, long-term follow-up has revealed issues such as progressively worsening aortic regurgitation (AR) and an increased frequency of repeated interventions [2, 3]. In contrast, surgical intervention appears to provide more durable long-term outcomes [4, 5]. Therefore, our center advocates for aortic valve (AoV) repair as the initial surgical strategy in infants, and children with congenital AS. Nevertheless, reports on the long-term outcomes of such patients remain limited, particularly regarding the characteristics of reinterventions.

How to optimize reinterventions in young patients to achieve a balance among long-term survival, quality of life, and the need for future reinterventions remains a critical challenge. The choice of reintervention strategy for AS generally depends on patient-specific factors, such as valve morphology and function [6]. Common surgical options include AoV re-repair, mechanical aortic valve replacement (mAVR), and the Ross procedure. The drawbacks of mAVR include the necessity for lifelong anticoagulation, patient–prosthesis mismatch, and elevated long-term risks [7]. Owing to its potential for growth, the Ross procedure is currently regarded as the preferred option for AoV replacement in children; however, it also entails the risk of multiple future replacements of the right ventricular–pulmonary artery (RV-PA) conduit [8]. To date, the strategies and outcomes of reinterventions following initial surgery for pediatric AS have not been thoroughly elucidated.

Therefore, we conducted a retrospective analysis of the long-term outcomes in children with isolated congenital AS who underwent initial AoV repair at a single center, with a specific focus on the treatment strategies and outcomes of those who required reintervention.

Methods

Patients

A retrospective analysis was conducted on patients with AS who were treated between January 2013 and December 2024. The inclusion criteria were as follows: (1) Diagnosis of AS confirmed by intraoperative findings, with assessment of transvalvular pressure gradient via echocardiography and evaluation of AoV morphology via cardiac computed tomography (CT); (2) Undergoing surgical AoV repair, including AoV commissurotomy and valvuloplasty; (3) Possible coexistence of simple congenital heart diseases such as mild coarctation of the aorta, ventricular septal defect, atrial septal defect, or patent ductus arteriosus. The exclusion criteria were as follows: (1) Concomitant moderate or greater AR; (2) Conversion to AoV balloon valvuloplasty, Ozaki procedure, Ross procedure, or mAVR after failed repair; (3) Non-congenital AS due to conditions such as infective endocarditis or rheumatic heart disease; (4) Previous AoV surgery, including arterial switch operation, Ross procedure, or mAVR; (5) Coexistence of complex congenital heart diseases such as tetralogy of Fallot, pulmonary atresia, hypoplastic left heart syndrome, or interrupted aortic arch [6].

The primary surgical indication for AS is peak AoV pressure gradient > 50 mmHg. For reoperation, the surgical indications include peak AoV pressure gradient > 50 mmHg or moderate-to-severe or greater AR. If accompanied by symptoms such as syncope, decreased exercise tolerance, or heart failure, even if the aforementioned hemodynamic thresholds are not met, surgical intervention is directly indicated.

Operative technique

The initial AoV repair was performed as previously reported, routinely via a median sternotomy with aortic–bicaval cannulation under mild hypothermic cardiopulmonary bypass. AoV repair was conducted through an oblique aortotomy, with the incision carefully limited along the aortic commissures to avoid extension into the aortic wall. Adequate leaflet coaptation was ensured through techniques such as leaflet thinning, patch augmentation, and commissural suspension [2].

During reoperation, the surgical approach was selected based on the condition of the AoV. If valve function could be preserved through repair techniques, a repeat repair was performed; otherwise, AoV replacement was undertaken. The replacement procedures were as follows: The Ross procedure involved harvesting the patient’s own pulmonary valve, which was implanted as a neo-aortic root, along with concomitant coronary reimplantation. Reconstruction of the RV-PA continuity was accomplished using a bovine jugular vein valved conduit (BalMedic, Beijing, China) or a handcrafted trileaflet expanded polytetrafluoroethylene (Gore-tex; Preclude, W. L. Gore & Associates, Inc, AZ, USA) valved conduit. mAVR was performed in cases with contraindications such as connective tissue disease, autoimmune diseases, coronary anomalies, pulmonary valve disease, or significant size discrepancy between the aortic and pulmonary annuli. mAVR involved excision of the native valve along the annulus, followed by suturing of an appropriately sized mechanical prosthesis into the aortic annulus. All AoV replacements performed before 2018 utilized mAVR.

Definitions

The primary outcome was the rate of reoperation after the initial procedure and secondary surgeries, while secondary outcomes included the rate of postoperative AoV replacement and the incidence of postoperative moderate or greater AS or AR. Follow-up was conducted until the occurrence of the primary outcome or the last follow-up date.

Valve morphology was described based on direct intraoperative observation and with reference to the Sievers classification. The Z-scores of preoperative aortic annulus diameter, left ventricular end-systolic dimension (LVESD), and left ventricular end-diastolic dimension (LVEDD) were calculated according to previously reported methods [9]. The initial AoV repair procedures were defined as follows: Aortic valvotomy referred to surgical AoV commissurotomy without concomitant repair techniques, while aortic valvuloplasty included surgical techniques such as leaflet thinning, leaflet patch repair, leaflet plication, triangular resection, and commissural annuloplasty. Postoperative complications included mortality and extracorporeal membrane oxygenation (ECMO) support. Early mortality was defined as death occurring within 30 days postoperatively or before hospital discharge. Moderate or greater AS or AR during postoperative follow-up was defined as a peak gradient ≥ 36 mmHg or a regurgitant jet width > 4 mm on echocardiography [10]. In the reoperation cohort, AoV replacement included the Ross procedure and mAVR.

Data analysis

Statistical analyses were performed using SPSS Statistics 22.0 (IBM, Chicago, IL) and GraphPad Prism 8 software (GraphPad Software, Inc). Continuous variables were assessed for normality using the Shapiro–Wilk test. They were described as mean ± standard deviation (SD) or median (interquartile range [IQR]) and were analyzed using Student’s t-test or the Mann–Whitney U test. Categorical variables were expressed as frequency (percentage), and intergroup comparisons were performed using the chi-square test or Fisher’s exact test. Time-to-event endpoints were analyzed and plotted using Kaplan–Meier actual survival curves, and differences in time-dependent endpoints between groups were assessed with the log-rank test. Univariate Cox proportional hazards models were used to identify risk factors for reoperation, AoV replacement, and moderate or greater AS/AR. Variables with a p-value < 0.1 in the univariate analysis were included in a forward stepwise multivariate analysis. Results are presented as hazard ratios (HR) with 95% confidence intervals (CI). A two-sided p-value < 0.05 was considered statistically significant.

Results

Patient characteristics

A total of 203 patients with AS were included in this study. Supplemental Table S1 presents the demographic and baseline characteristics of patients who underwent primary AoV repair. The cohort consisted of 133 males and 70 females, with median age of 2.4 (0.6, 4.7) years (range 9days–14years), including 71 (35.0%) infants and 132 (65.0%) children. Valve morphology comprised unicuspid AoV (n = 1, 0.5%), bicuspid AoV (n = 138, 68.0%), and tricuspid AoV (n = 64, 31.5%). The mean preoperative peak AoV pressure gradient was 83.5 ± 25.1 mmHg, and the mean left ventricular ejection fraction (LVEF) was 71.2 ± 11.6%. The mean aortic annulus diameter was 1.3 ± 0.9 cm, with a mean Z-score of 0.3 ± 1.5. The mean LVESD was 1.8 ± 0.5 cm, with a mean Z-score of −0.3 ± 2.3, and the mean LVEDD was 3.2 ± 1.5 cm, with a mean Z-score of 0.0 ± 1.9.

Perioperative results

In the study cohort, 133 patients (65.5%) underwent aortic valvotomy, while 70 patients (34.5%) were subjected to aortic valvuloplasty. The median cardiopulmonary bypass (CPB) time was 57.0 (43.0, 77.0) minutes, and the median aortic cross-clamp (ACC) time was 33.0 (20.0, 45.0) minutes. Postoperative transesophageal echocardiography revealed a mean peak AoV pressure gradient of 26.1 ± 14.8 mmHg. Immediate post-procedure transvalvular pressure gradient was ≤ 36 mmHg in 160(78.8%) patients, while it was > 36 mmHg in 43(21.2%) patients. Mild-to-moderate or greater AR was observed in 10 patients (4.9%). The median postoperative intensive care unit (ICU) stay was 2.0 (2.0, 4.0) days, and the median postoperative hospital stay was 8.0 (7.0, 11.0) days. Postoperative complications occurred in 7 patients (3.5%), including extracorporeal membrane oxygenation (ECMO) support in 6 patients (3.0%) and 1 case (0.5%) of early mortality. The deceased patient experienced recurrent ventricular fibrillation postoperatively, along with moderate-to-severe AR, and underwent mAVR on postoperative day 4. However, ventricular fibrillation recurred after reoperation, accompanied by hemodynamic instability, and treatment was withdrawn at the family’s request due to financial considerations. Detailed perioperative outcomes are presented in Supplemental Table S1.

Long-term outcomes

The mean follow-up duration was 5.2 ± 2.8 years, with no long-term mortality observed. At the final follow-up, the mean peak AoV pressure gradient was 43.5 ± 29.0 mmHg, and 12.3% (25/203) of patients had moderate or greater AR. A total of 23.6% (48/203) of pediatric patients underwent reoperation due to AoV-related issues. The indications for reoperation included AS (32/48; 66.7%), AR (8/48; 16.7%), and mixed lesions (8/48; 16.7%). Among patients who underwent reoperation, the preoperative mean peak AoV pressure gradient was 73.7 ± 36.5 mmHg, and 33.3% (16/48) of patients had moderate or greater AR. The reoperations included 20 cases of aortic valve re-repair, 8 Ross procedures, and 20 mAVR. Data of patients who underwent AoV replacement are presented in Supplemental Table S2.

The freedom from AoV reoperation rate was 85.5% (95% CI, 82.0%–91.0%) at 5 years and 50.9% (95% CI, 36.2%–65.6%) at 10 years (Fig. 1a). Univariate and multivariate Cox proportional hazards models of risk factors for AoV reoperation are summarized in Table 1. Significant risk factors identified included smaller preoperative aortic annulus Z-score (P < 0.001; HR = 0.675; 95% CI: 0.552–0.825), higher preoperative peak AoV gradient (P = 0.024; HR = 2.093; 95% CI: 1.102–3.976), and higher postoperative peak AoV gradient (P = 0.049; HR = 1.849; 95% CI: 1.001–3.416).

Fig. 1.

Overall freedom from AoV reoperation (a); overall freedom from AoV replacement (b)

Table 1.

Results of Cox proportional hazards models for AoV reoperation

	Univariate analysis		Multivariate analysis
Characteristics	P-value	HR (95%CI)	P-value	HR (95%CI)
Gender (male/female)	0.208	0.658 (0.343, 1.263)
Age (years)	0.242	1.051 (0.967, 1.143)
Weight (kg)	0.120	1.019 (0.995, 1.043)
AoV morphology (non-tricuspid/tricuspid)	0.081	1.693 (0.938, 3.055)
LVEF (%)	0.157	1.024 (0.991, 1.059)
AoV annulus z-score	< 0.001	0.653 (0.542, 0.786)	< 0.001	0.675 (0.552, 0.825)
LVESD z-score	0.186	0.914 (0.800, 1.044)
LVEDD z-score	0.256	0.918 (0.793, 1.064)
Preoperative AoV peak gradient (≤ 75/>75mmHg)	0.010	2.278 (1.217, 4.263)	0.024	2.093 (1.102, 3.976)
Surgical procedure (valvotomy/valvuloplasty)	0.004	2.362 (1.312, 4.252)
Postoperative AoV peak gradient (≤ 36/>36mmHg)	0.001	2.787 (1.551, 5.007)	0.049	1.849 (1.001, 3.416)
Postoperative mild-moderate or great AR	0.067	2.634 (0.933, 7.437)

Male sex, non-tricuspid AoV, preoperative AoV peak gradient ≤ 75mmHg, Postoperative AoV peak gradient ≤ 36mmHg, and aortic valvotomy served as the reference. P < 0.05 indicated in boldface

AoV Aortic valve, LVEF left ventricular ejection fraction, LVESD left ventricular end-systolic dimension, LVEDD left ventricular end-diastolic dimension, AR aortic regurgitation

The freedom from AoV replacement rate was 92.0% (95% CI, 87.7%–96.3%) at 5 years and 62.7% (95% CI, 46.0%–79.4%) at 10 years (Fig. 1b). Multivariate Cox proportional hazards models revealed that risk factors for AoV replacement included an aortic annulus diameter Z-score less than 0.5 (P = 0.014; HR = 0.220; 95% CI: 0.065–0.740), a preoperative peak AoV pressure gradient greater than 75 mmHg (P = 0.005; HR = 4.625; 95% CI: 1.593–13.427), and the presence of mild-to-moderate or greater AR postoperatively (P = 0.001; HR = 6.242; 95% CI: 20.76–18.767) (Fig. 2).

Fig. 2.

Freedom from aortic valve (AoV) replacement stratified by preoperative AoV peak gradient greater than 75 mmHg (a), AoV annulus diameter Z-score greater than 0.5 (b), and immediate postoperative presence of mild-moderate or greater aortic regurgitation (AR) (c)

Reoperations

The perioperative data of 48 patients undergoing reoperation were compared between the AoV repair group and the AoV replacement group, as detailed in Table 2. No significant differences were observed between the two groups in terms of gender or the interval from the initial surgery. Patients who underwent AoV replacement were significantly older (P = 0.022) and had greater body weight (P = 0.021). The AoV repair group predominantly presented with AS (n = 18, 90.0%), while moderate or greater AR was more frequent in the AoV replacement group (n = 14, 50.0%; P = 0.005). Preoperative LVEF (P = 0.040), aortic annulus diameter (P = 0.013), LVESD (P = 0.004), and LVEDD (P = 0.009) were significantly larger in the AoV repair group than in the AoV replacement group. Regarding intraoperative data, both CPB time (P < 0.001) and ACC time (P < 0.001) were significantly longer in the AoV replacement group. Postoperative echocardiography revealed that mild-to-moderate AS/AR was present in 5 patients (25.0%) in the AoV repair group, whereas only mild or lesser degrees of AS/AR were observed in the AoV replacement group (P = 0.009). Patients in the AoV replacement group exhibited longer durations of both ICU stay (P = 0.038) and hospital stay (P = 0.019). With respect to concomitant procedures, early mortality occurred in 3 patients (10.7%) in the AoV replacement group. One case was the aforementioned patient, and the other two patients, who developed postoperative circulatory instability and were subjected to ECMO support, ultimately experienced withdrawal of care after failure to maintain cardiac function.

Table 2.

Comparison of baseline characteristics and perioperative data between patients undergoing AoV repair versus AoV replacement during reoperation

Variables	Overall (n = 48)	AoV repair (n = 20)	AoV replacement (n = 28)	P-value
Gender (male)	32 (66.7)	11 (55.0)	21 (75.0)	0.147
Age (years)	8.5 ± 4.0	6.7 ± 3.6	9.4 ± 4.0	0.022
Weight (kg)	29.0 ± 14.1	22.8 ± 12.1	32.2 ± 14.3	0.021
Follow-up time (y)	5.4 ± 3.4	5.0 ± 3.3	5.6 ± 3.5	0.569
Indication for reoperation				0.018
AS	32 (66.7)	18 (90.0)	14 (50.0)
AR	8 (16.7)	1 (5.0)	7 (25.0)
AS/AR	8 (16.7)	1 (5.0)	7 (25.0)
AoV peak gradient (mmHg)	73.7 ± 36.5	79.5 ± 27.7	69.5 ± 41.7	0.355
Moderate or great AR	16 (33.3)	2 (10.0)	14 (50.0)	0.005
LVEF (%)	73.4 ± 9.2	76.7 ± 6.7	71.1 ± 10.2	0.040
AV annulus diameter	1.6 ± 0.4	1.4 ± 0.3	1.7 ± 0.5	0.013
AV annulus z-score	0.4 ± 2.0	0.3 ± 1.5	0.6 ± 2.3	0.596
LVESD	2.3 ± 6.3	2.0 ± 0.3	2.5 ± 0.7	0.004
LVESD z-score	−0.7 ± 1.9	−1.2 ± 1.3	−0.4 ± 2.2	0.155
LVEDD	3.9 ± 0.8	3.5 ± 0.4	4.1 ± 0.9	0.009
LVEDD z-score	−0.1 ± 1.8	−0.2 ± 1.5	0.0 ± 2.0	0.643
CPB time (min)	145.0 (72.5, 196.5)	69.5 (52.0, 141.5)	168.5 (130.5, 247.0)	< 0.001
ACC time (min)	99.5 (49.3, 127.5)	48.0 (26.5, 89.8)	123.0 (99.3, 162.3)	< 0.001
Post-bypass TEE				0.009
≤Mild AS/AR	43 (89.6)	15 (75.0)	28 (100.0)
Mild-moderate AS/AR	5 (10.4)	5 (25.0)	0
ICU stay (d)	3.5 (2.0, 6.5)	2.0 (1.3, 4.8)	4.0 (3.0, 7.0)	0.038
Postoperative hospital stay (d)	12.0 (7.3, 18.8)	8.0 (7.0, 15.5)	13.5 (11.0, 19.8)	0.019
Postoperative complications
Early death	3 (93.8)	0	3 (10.7)	0.255
ECMO	2 (4.2)	0	2 (7.1)	0.504

Data presented as mean ± SD, median (IQR) and n (%). P < 0.05 indicated in boldface

AoV Aortic valve, AS Aortic stenosis, AR Aortic regurgitation, LVEF Left ventricular ejection fraction, LVESD Left ventricular end-systolic dimension, LVEDD Left ventricular end-diastolic dimension, CPB Cardiopulmonary bypass, ACC Aortic cross-clamping, ICU Intensive care unit, ECMO Extracorporeal membrane oxygenation

Follow-up outcomes of reoperated patients

The mean follow-up duration was 4.6 ± 2.7 years, with no long-term mortality observed. No significant difference was found in the reoperation rate between the AoV repair group and the AoV replacement group. However, a higher proportion of patients in the AoV repair group developed moderate or greater AS/AR [12 (60.0%) vs. 3 (12.0%); P = 0.001] (Supplemental Table S3). Three patients required a third surgical intervention, one patient had undergone AoV re-repair during the second operation and developed severe AR during follow-up, while the other two patients, who had undergone mAVR, developed severe AS. Detailed information is presented in Supplemental Table S4. The overall freedom from reoperation rate at 5 years postoperatively was 94.6% (95% CI: 87.2%−100.0%), and the 10-year freedom from reoperation rate was 87.9% (95% CI: 73.4%−100.0%). The freedom from moderate or greater AS/AR rate was 60.4% (95% CI: 43.0%−77.8%) at 5 years postoperatively, which decreased to 50.2% (95% CI: 30.8%−69.6%) at 10 years postoperatively (Fig. 3).

Fig. 3.

Overall freedom from AoV reoperation (a) and overall freedom from moderate or greater AS/AR (b) in reoperation patients

In the AoV replacement group, the rate of freedom from moderate or greater AS/AR was 90.9% (95% CI: 78.8%−100.0%) at 5 years postoperatively, declining to 81.9% (95% CI: 61.7%−100.0%) at 10 years postoperatively. In contrast, the AoV repair group demonstrated a rate of freedom from moderate or greater AS/AR of only 30.1% (95% CI: 6.4%−53.8%) at 5 years (P = 0.001) (Fig. 4a). Univariate analysis indicated that age < 10 years and aortic valve leaflet morphology were not identified as high-risk factors for the development of moderate or greater AS/AR postoperatively (Fig. 4). Within the AoV replacement group, no statistically significant difference was observed in the risk of developing moderate or greater AS/AR between patients who underwent the Ross procedure and those who underwent mAVR (Fig. 4d).

Fig. 4.

Freedom from moderate or greater postoperative AS/AR stratified by AoV repair versus AoV replacement (a), surgical age greater than 10 Years (b), bicuspid versus tricuspid AoV (c), and Ross procedure versus mAVR (d)

Univariate and multivariate Cox proportional hazards models of risk factors for reoperation revealed that AoV re-repair (P = 0.029; HR = 0.219; 95% CI: 0.056–0.854) and the presence of immediate postoperative mild-to-moderate residual AS/AR (P = 0.020; HR = 4.419; 95% CI, 1.261–15.484) were associated with the development of moderate or greater AS/AR postoperatively (Table 3).

Table 3.

Results of Cox proportional hazards models for postoperative moderate or greater AS/AR in reoperation patients

	Univariate analysis		Multivariate analysis
Characteristics	P-value	HR (95%CI)	P-value	HR (95%CI)
Gender (male/female)	0.263	0.538 (0.182, 1.591)
Age (years)	0.071	0.884 (0.773, 1.011)
Weight (kg)	0.172	0.969 (0.926, 1.014)
Preoperative AoV peak gradient (≤ 75/>75mmHg)	0.452	1.480 (0.533, 4.111)
Preoperative mild-moderate or great AR	0.184	0.422 (0.119, 1.505)
LVEF	0.033	1.074 (1.006, 1.147)
AV annulus z-score	0.300	0.877 (0.683, 1.124)
LVESD z-score	0.042	0.752 (0.571, 0.990)
LVEDD z-score	0.204	0.829 (0.621, 1.107)
Indication for reoperation
AS/AR	Reference
AS	0.751	1.277 (0.282, 5.784)
AR	0.818	0.794 (0.111, 5.687)
Surgical procedure (AoV repair/replacement)	0.005	0.158 (0.044, 0.566)	0.029	0.219 (0.056, 0.854)
Postoperative mild-moderate or great AS/AR	< 0.001	8.556 (2.566, 28.527)	0.020	4.419 (1.261, 15.484)

Male sex, preoperative AoV peak gradient ≤ 75mmHg, and AoV repair served as the reference. P < 0.05 indicated in boldface

AoV Aortic valve, AS Aortic stenosis, AR Aortic regurgitation, LVEF Left ventricular ejection fraction, LVESD left ventricular end-systolic dimension, LVEDD Left ventricular end-diastolic dimension

Discussion

AS in children presents a lifelong management challenge. While early valve replacement carries risks and suboptimal durability [11, 12], our strategy prioritizes initial AoV repair to relieve stenosis and preserve the native valve whenever possible. This approach was associated with excellent early survival in our cohort, providing a crucial and safe bridge to later life. The risk of reoperation is assessed based on follow-up outcomes, with a preference for performing the second surgery after adolescence whenever possible. BVPL was not selected as the initial treatment plan because surgical AoV repair yields stable outcomes with a low incidence of early AR [13]. In contrast, BVPL may lead to significant AR in the early postoperative period, particularly in patients with bicuspid aortic valve [14, 15]. Our observed 5-year freedom from reoperation and replacement rates compare favorably with historical data, while the 10-year outcomes align with contemporary reports, supporting the clinical utility of our approach [6]. These findings indicate that our treatment strategy holds significant clinical applicability.

Consistent with prior studies, our analysis confirms that anatomic and hemodynamic factors—namely a small aortic annulus, high preoperative gradient, and suboptimal immediate postoperative result—are key determinants of reintervention risk [2]. The convergence of conclusions across different treatment modalities suggests that the primary determinant of prognosis is the congenital development of the AoV. In previous studies, surgery during infancy has also been considered a risk factor for reintervention [6]; however, this association was not observed in the present study. No early mortality occurred among the 69 (34.0%) infant patients, indicating that the treatment strategy of relieving stenosis to delay the timing of AoV replacement is effective. Therefore, patients with a small annulus or severe pressure gradients should be considered high-priority for close follow-up, as they face an elevated risk of requiring reintervention in the short term.

Following initial repair surgery for the relief of AS in children, a majority of patients may undergo a second operation in the future [16]. The decision-making for reintervention still requires comprehensive consideration of multiple factors, including patient characteristics, valve durability, long-term quality of life, anticoagulation regimen, reproductive intentions, and the values of both patients and their families [1]. With advances in surgical techniques and perioperative management, strategies for AoV repair continue to be refined. At our center, the indications for reoperation are determined based on the patient’s clinical symptoms and echocardiographic findings—specifically, a peak transvalvular pressure gradient > 50 mmHg or the presence of moderate-to-severe or greater AR. A key consideration is how the choice of AoV intervention during the second surgery can benefit the patient by reducing the likelihood of a third operation and improving their quality of life.

Our data support a selective approach to reoperation. The finding that re-repair was primarily performed for recurrent stenosis whereas replacement was favored for significant regurgitation reflects a deliberate selection strategy. This is based on the pathophysiology: pliable leaflets may be amenable to repeat repair, while chronic stress often leads to irreversible changes (prolapse, calcification) necessitating replacement. Historically, our center’s treatment strategy during secondary surgery prioritized aortic valve repair whenever feasible, based on the long-term durability of valve repair in individual patients. However, since the introduction of the Ross procedure in 2018, we have gradually adopted it as the preferred approach for secondary surgery, consistent with previous studies [17]. The mean follow-up duration after initial surgery was 5.4 years, during which high-risk AoV replacement during infancy was successfully avoided. Although three early postoperative deaths occurred, long-term outcomes indicated better quality of life in the AoV replacement group. More than half of the patients in the AoV repair group developed moderate or greater AS/AR within five years, which significantly impaired daily physical activity and imposed substantial psychological burdens on patients and their families [11, 18]. Based on these findings, Cox proportional hazards models further identified AoV repair and immediate postoperative mild-to-moderate or greater AR as risk factors for long-term moderate or greater AS/AR. Therefore, opting for AoV replacement during secondary surgery may represent a more favorable strategy.

The primary surgical options for AoV replacement are mAVR and the Ross procedure. Unless contraindications exist—such as connective tissue disease, autoimmune diseases, significant aortopulmonary annular discrepancy, or pulmonary valve pathology—our preferred approach remains the Ross procedure. Two patients who required a third operation had initially undergone mAVR and developed severe AS during follow-up, necessitating reoperation. Within the replacement cohort, no significant difference in medium-term freedom from AS was observed between Ross and mAVR, though this analysis is limited by sample size and follow-up duration. The choice between mAVR and the Ross procedure remains controversial [19]. While mAVR is associated with a slightly lower long-term rate of all-cause reintervention, its main drawbacks include the requirement for long-term anticoagulation and the risk of patient–prosthesis mismatch [7, 20]. In contrast, the Ross procedure eliminates the need for anticoagulation, thereby significantly improving patients’ quality of life and yielding long-term survival rates comparable to those of the general population [11]. The main challenge associated with the Ross procedure is the potential need for RV-PA conduit replacement [21–23]. Nonetheless, considering the benefits in terms of life expectancy and quality of life, the risk of conduit replacement appears acceptable [24]. Additionally, previous studies have reported superior outcomes with the second-stage Ross procedure compared to the primary operation [17]. This may be attributed to adhesions formed after the initial sternotomy, which can provide additional external support to the autograft [17]. These findings further support our treatment strategy.

Limitations

As a retrospective study, this investigation inevitably has several limitations. First, the follow-up duration was relatively short; although the proportion of patients lost to follow-up was small, it could still be a source of calculation bias. Second, differences in baseline characteristics between the mAVR and Ross groups among patients who underwent reoperation may affect the reliability of the conclusions. However, using propensity score matching to reduce such differences would also be constrained by the limited sample size, which could similarly influence the conclusions drawn. Third, the study cohort included infants and children, whose disease spectrum and prognosis may differ. Although age was adjusted for in the multivariate analysis, combining different subgroups for analysis may obscure risk factors specific to particular populations. Future studies with larger sample sizes are needed to perform subgroup analyses. Finally, as a single-center study initiated in 2018, the relatively short follow-up period may be compounded by the potential impact of the learning curve associated with the Ross procedure. Prospective, multicenter studies with long-term follow-up are needed to further clarify the optimal treatment strategy for reoperation following aortic valve–sparing surgery.

Conclusions

Patients with congenital AS can achieve favorable therapeutic outcomes after undergoing initial AoV repair. However, nearly half of these patients may require reoperation within 10 years. During reoperation, AoV replacement, particularly the Ross procedure, may be more likely to yield desirable long-term quality of life.

Abbreviations

AS

Aortic stenosis

AoV

Aortic valve

BVPL

Balloon valvuloplasty

AR

Aortic regurgitation

mAVR

Mechanical aortic valve replacement

RV-PA

Right ventricular–pulmonary artery

CT

Computed tomography

LVESD

Left ventricular end-systolic dimension

LVEDD

Left ventricular end-diastolic dimension

ECMO

Extracorporeal membrane oxygenation

CPB

Cardiopulmonary bypass

ACC

Aortic cross-clamp

Authors’ contributions

J.Z. and X.H. contributed to the study conception and design. Material preparation, data collection and analysis were performed by G.Z., XJ.Z., L.H., B.S., K.L., XY.Z. and H.L. The first draft of the manuscript was written by G.Z. and XJ.Z. All authors gave final approval and agreed to be accountable for all aspects of work ensuring integrity and accuracy.

Funding

This work was supported by the National Natural Science Foundation of China (grant number 82172101), the Natural Foundation of Shanghai Science and Technology Committee (grant number 23Y11907000, 23ZR1440900), Shanghai Key Research Center Construction Project-Shanghai Research Center for Pediatric Cardiovascular Diseases (2023ZZ02024), National Clinical Key Specialty Construction Project (10000015Z155080000004).

Data availability

Data supporting the study conclusions can be obtained from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Human Research Ethics Committee of the Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine (SCMC1RB—K2022171—1). All consent to participate were waived by the Human Research Ethics Committee of the Shanghai Children’s Medical Center due to the retrospective nature of the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.