Treatment of valvular aortic stenosis in children: a 20-year experience in a single institution

Transcription

1 Interactive CardioVascular and Thoracic Surgery 27 (2018) doi: /icvts/ivy078 Advance Access publication 19 March 2018 ORIGINAL ARTICLE Cite this article as: Kjellberg Olofsson C, Berggren H, Söderberg B, Sunnegårdh J. Treatment of valvular aortic stenosis in children: a 20-year experience in a single institution. Interact CardioVasc Thorac Surg 2018;27: Treatment of valvular aortic stenosis in children: a 20-year experience in a single institution Cecilia Kjellberg Olofsson a,b, *, Håkan Berggren a,björn Söderberg a and Jan Sunnegårdh a a b Department of Cardiology, The Queen Silvia Children s Hospital, Sahlgrenska University Hospital, Institute of Clinical Sciences, Gothenburg University, Gothenburg, Sweden Department of Pediatrics, Sundsvall Hospital, Sundsvall, Sweden * Corresponding author. Department of Pediatrics, Sundsvall Hospital, Sundsvall, Sweden. Tel: ; cecilia.kjellberg.olofsson@lvn.se (C. Kjellberg Olofsson). Received 8 November 2017; received in revised form 31 January 2018; accepted 16 February 2018 Abstract OBJECTIVES: This study presents short- and long-term follow-up after treatment for isolated valvular aortic stenosis in children with surgical valvotomy as the preferred 1st intervention. METHODS: All patients aged 0 18 years treated between 1994 and 2013 at our centre were reviewed regarding the mode of first treatment, mortality, reinterventions and the need for aortic valve replacement. RESULTS: A total of 113 patients were identified in local registries. There were 44 neonates, 31 infants and 38 children. The mean followup period was 11 years (range 2 22 years). No early deaths and only 2 late deaths were reported. Of the 113 patients, 92 patients had open surgical valvotomy as the 1st intervention. Freedom from reintervention was 80%, 69%, 61%, 57% and 56% at 1, 5, 10, 15 and 20 years, respectively. The main indication for reintervention was valvular stenosis. Freedom from aortic valve replacement was 67%. CONCLUSIONS: Surgical valvotomy of aortic stenosis in this long-term follow-up study resulted in no 30-day mortality and <1% late mortality. Reinterventions were common, with 38% of the patients having surgery or catheter treatment of the aortic valve before the age of 18 years. Among the 40 patients aged 18 years or older at follow-up, 45% had had the aortic valve replaced. Our data do not allow comparison of catheter and surgical treatment, but, based on these results, we find no reason to change our current policy of surgical treatment as 1st intervention in patients with isolated valvular aortic stenosis. Keywords: Congenital valvular aortic stenosis Surgical valvotomy Outcome INTRODUCTION Congenital valvular aortic stenosis, especially critical neonatal aortic stenosis, is a complex and lifelong disease. Available treatment methods are associated with both advantages and disadvantages. Intervention in early life aims at preserving the native valve and delaying the need for aortic valve replacement (AVR) which, in growing children, is more frequently associated with suboptimal results [1]. Both surgical valvotomy (SAV) and balloon valvotomy (BAV) are firmly established as initial treatment methods, and there is an ongoing debate about which method is the most beneficial intervention during childhood [2 10]. There are also groups advocating early valve replacement as the best intervention [11], but no randomized series for different treatment strategies exist. Institutional preferences dictate the mode of primary intervention and later reinterventions. Using different treatment methods, outcomes have varied regarding early and late mortality and the need for reinterventions [3, 4, 12 14]. Since 1994, the Queen Silvia Children s Hospital in Gothenburg has been 1 of 2 centres for paediatric heart surgery in Sweden, serving approximately half of the Swedish paediatric population. In this study, we review 20 years of experience at our centre in treating isolated congenital aortic valve stenosis with SAV as the preferred 1st intervention. MATERIALS AND METHODS We reviewed all patients who had primary treatment for isolated valvular aortic stenosis at our institution between January 1994 and December Patients were identified in our local surgical and catheter registries and followed up until 18 years of age. Only patients selected for biventricular repair were included. Data were collected from patient files in our hospital as well as the referring hospitals. Transthoracic echocardiography was used for evaluating anatomy, Doppler-derived gradients, aortic valve regurgitation and left ventricular function. Doppler gradients were derived using the modified Bernoulli equation for the maximal gradient. Mean gradient was calculated from the traced velocity curve. Aortic VC The Author(s) Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 C. Kjellberg Olofsson et al. / Interactive CardioVascular and Thoracic Surgery 411 regurgitation was graded 0 4 based on the colour Doppler jet and retrograde flow in the descending aorta. For assessment of left ventricular function, left ventricular end-diastolic diameter and left ventricular fractional shortening were used. Hypertrophy was defined as z-scores for interventricular septum or left ventricular posterior wall in diastole >2 SD [15, 16]. Treatment criteria were critical neonatal aortic stenosis and Doppler mean gradient or invasive gradient >50 mmhg. If symptoms of chest pain or fatigue, ischaemic electrocardiographic (ECG) changes, pathological blood pressure reaction on bicycle ergometry testing, left ventricular hypertrophy or depressed left ventricular function were present, a lower gradient was accepted as an indication of intervention. Critical aortic stenosis in the neonate was defined as duct-dependent systemic circulation and/or depressed left ventricular function with left ventricular fractional shortening <28% [17]. Only patients with isolated valvular aortic stenosis were included. Patients with additional heart defects requiring interventional or surgical treatment were excluded. Survival was cross-checked against the Swedish Population Registry as of October The preferred treatment option in our institution is SAV, i.e. commissurotomy including thinning of dysplastic leaflets and shaving off noduli when appropriate. Statistical analysis SPSS statistical software was used for data analysis. Categorical variables were reported as absolute numbers and percentages. Continuous variables were expressed as either mean ± standard deviation or median value and range. For all tests, a P-value <0.05 was considered statistically significant. Freedom from reintervention was calculated using the Kaplan Meier method. This study was approved by the Regional Ethics Review Board of Western Sweden (approval no ). RESULTS Figure 1 shows all interventions and status of 113 patients who met the inclusion criteria for biventricular repair. At 1st intervention, 44 patients were neonates (<_30 days old), 31 patients were infants (1-month to 1-year-old) and 38 patients were children (>1 year). Critical aortic stenosis was diagnosed in 28 of the neonates, of whom 10 had symptoms of low cardiac output at presentation. Five neonates with critical aortic stenosis, aortic valve annulus <5 mm and borderline left ventricle assessed as not adequate to sustain systemic circulation were excluded as they underwent univentricular palliation. A 99 patients were diagnosed with hypoplastic left heart syndrome. The median age at diagnosis was 2 days (range 0 days to 13.6 years; girls 28% and boys 72%). The median age at 1st intervention was 2.8 months ( years). No differences were observed between boys and girls regarding gestational age, birth weight or age at diagnosis (Table 1). The only dropouts were 3 patients who emigrated at 3, 13 and 14 years of age (3, 11 and 14 years after 1st intervention). The median follow-up time was 11.2 years (range years) for mortality and 9 years (range 2 18 years) for reinterventions. Echocardiographic and catheter-derived data are presented in Table 2. Assessment of maximal and mean Doppler gradient before and after SAV or BAV showed a significant decrease from 101 ± 27 to 49 ± 19 mmhg (maximal gradient) and from 57 ± 16 to 29 ± 14 mmhg (mean gradient) for the whole group. Doppler gradients showed a decrease in the SAV group from 104 ± 25 to 47 ± 16 mmhg (maximal gradient) and from 58 ± 15 to 28 ± 12 mmhg (mean gradient) and from 89 ± 39 mmhg to 67 ± 27 mmhg (maximal gradient) and from 56 ± 23 mmhg to 39 ± 21 mmhg (mean gradient) in the BAV group (Fig. 2). Regression analysis demonstrated a statistically significant relationship between left ventricular end-diastolic diameter z-score, Doppler mean gradient (r = -0.32, P = 0.004) and maximal gradient (r = -0.28, P = 0.007) and between interventricular septum in diastole z-score, Doppler mean gradient (r = 0.25, P = 0.033) and maximal gradient (r = 0.27, P = 0.015). No significant relationship was observed between left ventricular posterior wall in diastole and either the mean or the maximal Doppler gradient across the aortic ostium. Comparison of the Doppler mean gradient and catheterderived peak-to-peak gradient was possible in 38 cases. A statistically significant correlation was noted between the catheterderived gradient (mean 55 ± 20 mmhg) and the Doppler mean gradient (mean 53 ± 16 mmhg) (r = 0.54, P < 0.000). Of the initial procedures, there were 92 SAVs, 11 BAVs, 4 Ross procedures, 2 closed transapical valve dilatations and 4 prosthetic valve replacements (Table 3). The primary BAVs were carried out mainly between 2000 and 2006, 8 in the neonatal group, including 6 cases classified as critical valvular aortic stenosis. Three patients were older than 1 year at BAV, of whom 2 had complex extracardiac diseases. Two neonates and 2 children had primary Ross procedures. The 2 closed transapical valve dilatations were carried out in the early study period, 1 in a premature baby weighing 1.5 kg and the other in a baby born at term but small for gestational age, weighing 2.2 kg. No 30-day mortality was reported. One late death occurred in a 10-month-old boy with critical aortic stenosis and severe endocardial fibroelastosis of the left ventricle who presented with severe heart failure. He underwent BAV and a repeat BAV, followed by SAV during the first 2 months of life and was later diagnosed with mitral valve stenosis and died due to heart failure and concomitant pulmonary infection. The other late death was a boy with previous BAV at 13 years of age with prior kidney transplantation, developmental delay and finger malformation. He received a mechanical prosthesis 15 years after the initial BAV treatment and had postoperative left ventricular failure followed by an emergency heart transplantation. He succumbed due to graft failure in the postoperative period at 28 years of age. Reintervention, regardless of the type of initial intervention, was required in 48 (43%) patients at a median age of 4.3 years ( years) and a median 1.2 years (1 day to 16.7 years) after the 1st intervention. Freedom from reintervention was 80%, 69%, 61%, 57% and 56% at 1, 5, 10, 15 and 20 years, respectively (Fig. 3). Reintervention of the aortic valve was performed in 43 patients (38% of the whole cohort), including 58% of the neonates. No significant difference in time was observed from the 1st intervention to reintervention in the different age groups. Residual stenosis was the most common reason for the 1st reintervention, and, on that indication, BAV was the method most commonly used (Fig. 4). The method of reintervention was decided on a case-by-case basis (Table 3). No patient had a significant (Grade 3 or 4) aortic regurgitation at discharge after the 1st intervention. Grade 2 aortic regurgitation was seen in 11 cases, 10 of the 93 after SAV and 1 of the 11 after BAV (n.s.). CONGENITAL

3 412 C. Kjellberg Olofsson et al. / Interactive CardioVascular and Thoracic Surgery All pa ents n=113 neonate (neo): 44 infant(inf): 31 children (ch): 38 UVH HTX Death HTX 1st interven on surgery n=102 SAV: 92 (58) CTV: 2 (0) neo: 36 Ross: 4 (4) inf: 31 AVR Mec: 3 (3) ch: 35 AVR Biol: 1 (0) 21 2nd interven on surgery n=25 SAV: 5 (1) Ross:6 (6) 3 AVR Mec: 7(7) inf: 3 AVR Bio: 3 (0) ch: 9 AVR Hom: 4 (1) 3rd Interven on surgery n=16 SAV: 4 (1) Ross: 5 (4) neo: 9 AVR Mec: 3 (3) inf: 4 AVR Bio: 2 (2) ch: 3 AVR Hom: 2 (1) 4th interven on surgery n=5 AVR Mec: 3 (3 ) AVR Bio: 2 (1 ) neo:3 inf:2 Two patients had heart transplantation at the age of 1.5 and 15 years, respectively. One patient, with a well-developed left ventricle and aortic arch of adequate size but almost atretic aortic valve, had SAV but this was converted on the 1st postoperative day to Norwood surgery when the mitral valve showed severe dysplasia and very poor left ventricular function was seen. AVR was performed in 34 (33%) patients (Table 3). Freedom from AVR at 18 years of age was 56%. Three of the 18-year-olds were born with critical aortic stenosis, and they all had AVR before 18 years of age. A follow-up as of 31 December 2016 revealed that an additional 7 patients had had an AVR. Syndromes or birth defects were noted in 8 patients at admission: Alagille syndrome, pyloric stenosis, pes equinovarus, congenital hip dysplasia, vertebral malformation, renal dysfunction, Turner syndrome and hypospadia. Alivewithout 65 interven on 4 n=69 Alivewithout interven on n=24 Alive without 11 reinterven on 1 n=12 Alivewithout reinterven on n=4 1st interven on BAV n= neo: 8 inf: 0 ch: 3 2nd interven on BAV n= inf:3 ch: 4 3rd interven on BAV n= 2 Figure 1: Flow chart of all interventions and status at follow-up of all patients. Numbers within parenthesis indicate patients with no intervention on aortic valve. AVR: aortic valve replacement; BAV: balloon valvotomy; Bio: biological prosthesis; CTV: closed transventricular valvotomy; Hom: homograft; HTX: heart transplant; Mec: mechanical prosthesis; Ross: Ross procedure; SAV: surgical valvotomy; UVH: univentricular heart. inf: 0 ch: 1 Concomitant structural heart disease diagnoses considered insignificant at the initial treatment were mild aortic regurgitation, mildto-moderate mitral regurgitation, insignificant ventricular septal defects or atrial septal defects, persistent ductus arteriosus and left superior vena cava. In 11 neonates with critical aortic stenosis, endocardial fibroelastosis was found in mild-to-moderate form and in 4 neonates a severe endocardial fibroelastosis was found. Valve morphology was established from surgical reports. The valve was described as functionally unicuspid (n = 3), morphologically bicuspid (n = 12), functionally bicuspid (n =68), tricuspid (n = 20) and quadricuspid (n =1). Description of valve morphology was missing in 3 cases who had surgical interventionandin6caseswhohadonlybav.nostatisticallysignificant difference of the time to reintervention was found when testing all combinations of the different valve morphologies against the different age groups and SAV or BAV as the initial treatment.

4 C. Kjellberg Olofsson et al. / Interactive CardioVascular and Thoracic Surgery 413 Table 1: Patient characteristics Total Neonates (<_ 30 days) Infants (1 month to 1 year) Children (1 18 years) P-value Patients Gestational age (weeks), median (range) 40 (30 42) 39 (32 42) 40 (32 42) 40 (30 41) 0.69 Age at diagnosis (days), median (range) 2 (0 4875) 1 (0 20) 3 (0 183) 30 (0 4875) Age at 1st intervention, median (range) 2.8 months (0 6546) days 6.5 (0 26) days 3.6 ( ) months 12.8 ( ) years Gender, male/female 88/25 35/9 24/7 29/ Weight at birth (kg), mean (SD) 3.4 (0.8) 3.5 (4) 3.7 (2) 3.4 (3.3) 0.55 Weight at 1st intervention (kg), mean (SD) 19.4 (24) 3.5 (0.8) 6.5 (1.8) 48.1 (21) Follow-up time (years), mean (SD) 11 (5.8) 10 (5.4) 11 (6.1) 12 (13) 0.25 Kruskal Wallis test was used for difference between subjects. SD: standard deviation. Table 2: DISCUSSION Echocardiographic and catheter-derived data before and after primary intervention The present study includes all patients from a well-defined geographical area treated for significant valvular aortic stenosis.our institution is a tertiary centre for paediatric cardiac surgery, serving half of Sweden. The inflow of patients has been very stable over time since centralization of paediatric cardiac surgery in Sweden 1994, as no private option for such treatment is available [18]. Congenital valvular aortic stenosis is a fatal disease, immediate in newborns with critical aortic stenosis and leading to premature death in cases with a more moderate degree of stenosis. Thus, the aim of the treatment in the former group was life-saving and in the latter group to prevent untimely death. A main finding in the present study was low mortality, with no 30-day mortality and only 2 late deaths. Total (n = 113) Neonates (n = 44) Infants (n = 31) Children (n = 38) P-value LVEDd (z-score) -0.2 (1.9) (2.1) 0.12 (1.6) (1.9) n.s. IVSd (cm) 0.7 (0.3) (n = 79) 0.5 (0.1) (n = 24) 0.6 (0.2) (n = 25) 1.0 (0.3) (n = 30) *** IVSd (z-score) 1.5 (1.3) 1.2 (1.1) 1.7 (1.4) 1.6 (1.2) n.s. LVEDd (cm) 2.9 (1.3) (n = 92) 1.9 (0.5) (n = 33) 2.4 (0.5) (n = 26) 4.3 (0.9) (n = 33) *** LVPWd (cm) 0.7 (0.5) (n = 82) 0.6 (0.7) (n = 25) 0.5 (0.1) (n = 25) 0.9 (0.3) (n = 32) ** LVPWd (z-score) 1.6 (1.4) 1.7 (1.6) 1.7 (1.1) 1.4 (1.59 n.s. LVFS (%) 38 (12) (n = 90) 30 (12) (n = 33) 40 (9.0) (n = 25) 44 (8.3) (n = 32) *** Aortic annulus (cm) 1.2 (0.6) (n = 87) 0.7 (0.1) (n = 34) 0.9 (0.2) (n = 23) 2.0 (0.4) (n = 30) *** Aortic annulus (z-score) 0.09 (1.6) (1.6) 0.01 (1.6) 0.83 (1.2) ** Peak aortic gradient (mmhg) 97 (31) (n = 112) 86 (37) (n = 43) 113 (23) (n = 31) 95 (21) (n = 38) ** Residual peak aortic gradient (mmhg) 45 (22) (n = 110) 50 (24) (n = 43) 37 (12) (n = 31) 47 (25) (n = 36) * Mean aortic gradient (mmhg) 55 (16) (n = 98) 53 (20) (n = 34) 62 (14) (n = 27) 52 (13) (n = 37) * Residual mean aortic gradient (mmhg) 29 (14) (n = 59) 32 (14) (n = 27) 20 (6.1) (n = 13) 30 (15) (n = 19) * Peak aortic jet velocity (m/s) 4.8 (0.8) (n = 112) 4.5 (1.1) (n = 43) 5.3 (0.5) (n = 31) 4.8 (0.6) (n = 38) ** Residual peak aortic jet velocity (m/s) 3.3 (0.8) (n = 110) 3.4 (0.9) (n = 43) 3.0 (0.5) (n = 31) 3.3 (0.9) (n = 36) n.s. Catheter gradient at BAV (mmhg) 63 (33) (n = 11) 65 (36) (n = 8) 57 (29) (n = 3) n.s. Residual catheter gradient after BAV (mmhg) 30 (16) (n = 11) 34 (18) (n = 8) 20 (19) (n = 3) n.s. Values are presented as mean (standard deviation). Kruskal Wallis test was used of difference between subjects. Values in boldface and/or underlined text indicate difference between groups. n indicates number of patients with documented measurement. *P < **P < ***P <_ BAV: balloon valvotomy; IVSd: interventricular septum in end-diastole; LVEDd: left ventricular end-diastolic dimension; LVFS: left ventricular fractional shortening; LVPWd: left ventricular posterior wall end-diastole; n.s.: not significant. In a comprehensive systematic review and meta-analysis of the outcome after treatment of congenital valvular aortic stenosis in 2368 patients by Hill et al. [4], a 30-day mortality of 4% after SAV and 6% after BAV was found. Long-term survival at 10 years after the 1st intervention with SAV or BAV was 90% and 87%, respectively, whereas our study showed a 10-year survival rate after the 1st intervention of 99%. Long-term results after BAV in a Belgian study of 93 patients reported by Soulatges et al. [19] showed a total mortality of 27% in neonates with critical aortic stenosis and 11.8% in the whole cohort with a mean follow-up of 11.7 years. Ewert et al. [7] reported a 30-day mortality of 3%, and in the neonatal group 9%, after BAV in a multicentre study including 1004 patient. Data on long time survival was not presented. In Finland, in a nationwide survey after BAV, Kallio et al. [20] CONGENITAL

5 414 C. Kjellberg Olofsson et al. / Interactive CardioVascular and Thoracic Surgery Figure 2: Transvalvular gradient before and after 1st intervention with SAV or BAV. BAV: balloon valvotomy; SAV: surgical valvotomy. Table 3: Number of primary interventions and reinterventions Procedure Primary intervention Surgical valvotomy 92 8 Balloon valvotomy Closed transventricular dilatation 2 Ross procedure 4 11 Mechanical prosthesis 3 14 Biological prosthesis 1 6 Homograft in the aortic position 5 Univentricular heart surgery 1 Heart transplantation 2 Other a 20 Total Reinterventions a Patient ductus arteriosus, subvalvular aortic stenosis, pacemaker, endocarditis in aortic homograft, mitral valve regurgitation, stenosis in pulmonary homograft, univentricular heart surgery or heart transplantation due to diastolic dysfunction and thrombus in left atrial appendage. reported a mortality of 6.3% before hospital discharge and no late mortality at a median 6.9-years follow-up. Siddiqui et al. [3], in an Australian study, compared outcome after SAV or BAV in neonates and infants. They reported a 3% hospital mortality and 10% late mortality in the whole group with a mean follow-up period of 10 years. Another recently published article by Galoin-Bertail et al. [21] presented follow-up data from France on 83 patients aged <4 months treated with SAV reporting a 6% early mortality and 85% survival at 15 years with a median follow-up of 4.2 years. Figure 3: Freedom from reintervention in years by age group at 1st intervention. A log-rank test was run to determine whether there were differences in the distribution of the different age groups at reintervention. The distributions for the 3 age categories were not statistically significant: v 2 = 0.102, P = Figure 4: Indications for reinterventions. Asterisk indicates patent ductus arteriosus, subvalvular aortic stenosis, pacemaker, endocarditis in aortic homograft, mitral valve regurgitation, stenosis in pulmonary homograft, univentricular heart surgery or heart transplantation due to diastolic dysfunction and thrombus in left atrial appendage. Even though our mortality results compare favourably, mortality rates should be compared with caution, especially for early mortality, as they may reflect differences in patient selection. In the present study, 28 (25%) patients were classified as having a critical aortic stenosis. An additional 5 patients with critical aortic stenosis were considered unsuitable for biventricular repair and were thus excluded. One patient was converted to univentricular palliation on the 1st postoperative day after SAV, and 2 patients eventually had a heart transplantation. Few studies describe the features of borderline cases with critical aortic stenosis chosen for univentricular palliation, which is important when comparing outcome [22 24]. In our data, borderline cases constituted 5% of all patients undergoing univentricular heart surgery due to critical aortic stenosis or hypoplastic left heart syndrome between 1994

6 C. Kjellberg Olofsson et al. / Interactive CardioVascular and Thoracic Surgery 415 and This study includes patients with SAV, BAV or AVR as the initial treatment, with SAV being the preferred intervention. Nonetheless, 11 patients in our cohort had BAV as the initial treatment, in some cases due to a presumed lower risk of the procedure in patients with extracardiac morbidity. The small number of patients who had BAV as the primary treatment precludes a comparison with SAV in the present study. Adequate gradient reduction was achieved both with SAV and BAV, although more radically after SAV. No patient had significant aortic valve regurgitation at discharge. Other groups have described a higher degree of aortic regurgitation both after BAV and SAV, but in our study, we found Grade 2 aortic regurgitation in only 10% of the patients after the 1st treatment and no patient had severe aortic regurgitation (Grade 3 4) [13, 14]. This is consistent with a careful surgical strategy with commissurotomy and trimming of leaflets to relieve stenosis without risking severe regurgitation. Hraska et al. [6] argued that SAV is preferable, as it can actually improve functioning and lifespan of the native valve. In addition, some studies have demonstrated longer freedom from reintervention for SAV than BAV [3, 13]. In the present study, we found even longer freedom from reintervention: 80%, 69%, 61%, 57% and 56% at 1, 5, 10, 15 and 20 years, respectively. However, our finding of 58% reintervention in the neonatal subgroup at a mean age of 11.2 years matches the 10-year freedom from reintervention of 55% found by Siddiqui et al. [3] in a group of neonates and infants aged <1 year whose primary intervention was SAV or BAV. The most common indication for reintervention in our study was residual stenosis or restenosis. As stenosis, and not regurgitation, was the most common indication, a reintervention with BAV was the method most commonly used. This strategy, with SAV or BAV followed by a second BAV if needed, has been shown to delay the need for an AVR [25]. Furthermore, we did not find any association between age at initial intervention and time to the need for reintervention, which is contrary to the findings in studies using BAV as the primary intervention [8, 19, 26] but consistent with studies using SAV as the primary intervention [3, 21]. A major issue in patients with valvular aortic stenosis is the timing of AVR. Freedom from AVR was 67% in the present study but 50% in the subgroup of neonates with critical stenosis (n = 15). The Ross procedure (n = 15) and mechanical prosthesis (n = 14) were equally common but, as expected, the age distribution was different, with younger patients in the Ross group and older patients in the mechanical prosthesis group. Alternative AVR with biological prosthesis or homograft in the aortic position were relatively rare procedures (n = 5). In the present study, 40 patients reached 18 years of age at follow-up, and freedom from AVR in this subgroup was 55%. However, the 3 patients born with critical aortic stenosis all underwent AVR, as had an additional 7 patients in this specific subgroup of 40 patients when followed up again in 2016 (December 31). We do not have sufficient data to compare the need for AVR after SAV and BAV treatment. Freedom from AVR is an important aspect for followup of valvular aortic stenosis. Our finding of 67% freedom from AVR is substantially higher than the 39% freedom from AVR at 15 years after SAV presented by Galoin-Bertail et al. [21] who found a generally higher incidence of reinterventions. It is also slightly higher than the 57% freedom from AVR after BAV treatment reported by Brown et al. [8] but similar to the 70% after SAV treatment reported by Brown et al. [5] in a 20-year followup. In the meta-analysis by Hill et al. [4], freedom from AVR at 10 years was 76% for BAV and 81% for SAV. Treatment criteria for intervention or reintervention in children with non-critical valvular aortic stenosis, such as transvalvular gradients, regurgitation fraction or signs of impaired left ventricular function, is rarely reported or discussed in the literature. Analysis of the preoperative investigation in the present study revealed that 64 of the 137 eligible ECG recordings (47%), from both primary interventions and reinterventions, showed signs of left ventricular hypertrophy and/or ST-segment depression. Bicycle ergometry testing revealed ST-segment depression and/ or pathological blood pressure reaction in 29 of 49 (59%) cases in patients with non-critical valvular stenosis at primary intervention or at reintervention due to stenosis. Strengths and limitations The strength of this study is that it includes all patients treated for valvular aortic stenosis from a specific geographical area, corresponding to half of the Swedish population, as patients are divided between the 2 Swedish centres based on geography rather than on patient preferences or nature of medical condition. We have only included patients with isolated congenital valvular aortic stenosis. Only 3 patients were lost to follow-up, all due to emigration. A weakness is the retrospective design. Also, different inclusion criteria hinder direct comparison between studies. CONCLUSION In conclusion, SAV was performed as the primary intervention for isolated aortic stenosis in this long-term follow-up study, resulting in no 30-day mortality and <1% late mortality. Reinterventions were common, with 38% of the patients having intervention by surgery or catheter treatment of the aortic valve before the age of 18 years. No correlation between the age at 1st intervention and the time to reintervention was found. The main indication of 1st reintervention was stenosis of the valve. Among the patients who had reached the age of 18 years (n = 40) at the end of the study, 45% had the aortic valve replaced. Our data do not allow comparison of catheter and surgical treatment, but based on these results, we find no reason to change our current policy of surgical treatment as the 1st intervention in patients with isolated valvular aortic stenosis. Funding This work was supported by Sahlgrenska University Hospital, Gothenburg, Sweden. Conflict of interest: none declared. REFERENCES [1] Etnel JRG, Elmont LC, Ertekin E, Mokhles MM, Heuvelman HJ, Roos- Hesselink JW et al. Outcome after aortic valve replacement in children: a systematic review and meta-analysis. J Thorac Cardiovasc Surg 2016;151: e3. [2] Rehnström P, Malm T, Jögi P, Fernlund E, Winberg P, Johansson J et al. Outcome of surgical commissurotomy for aortic valve stenosis in early infancy. Ann Thorac Surg 2007;84: CONGENITAL

7 416 C. Kjellberg Olofsson et al. / Interactive CardioVascular and Thoracic Surgery [3] Siddiqui J, Brizard CP, Galati JC, Iyengar AJ, Hutchinson D, Konstantinov IE et al. Surgical valvotomy and repair for neonatal and infant congenital aortic stenosis achieves better results than interventional catheterization. J Am Coll Cardiol 2013;62: [4] Hill GD, Ginde S, Rios R, Frommelt PC, Hill KD. Surgical Valvotomy Versus Balloon Valvuloplasty for Congenital Aortic Valve Stenosis: A Systematic Review and Meta-Analysis. J Am Heart Assoc. 2016;5,doi: [5] Brown JW, Ruzmetov M, Vijay P, Rodefeld MD, Turrentine MW. Surgery for aortic stenosis in children: a 40-year experience. Ann Thorac Surg 2003;76: [6] Hraska V, Sinzobahamvya N, Haun C, Photiadis J, Arenz C, Schneider M et al. The long-term outcome of open valvotomy for critical aortic stenosis in neonates. Ann Thorac Surg 2012;94: [7] Ewert P, Bertram H, Breuer J, Dähnert I, Dittrich S, Eicken A et al. Balloon valvuloplasty in the treatment of congenital aortic valve stenosis a retrospective multicenter survey of more than 1000 patients. Int J Cardiol 2011;149: [8] Brown DW, Dipilato AE, Chong EC, Lock JE, McElhinney DB. Aortic valve reinterventions after balloon aortic valvuloplasty for congenital aortic stenosis: intermediate and late follow-up. J Am Coll Cardiol 2010;56: [9] Reich O, Tax P, Marek J, Rázek V, Gilík J, Tomek V et al. Long term results of percutaneous balloon valvoplasty of congenital aortic stenosis: independent predictors of outcome. Heart 2004;90:70 6. [10] Miyamoto T, Sinzobahamvya N, Wetter J, Kallenberg R, Brecher AM, Asfour B et al. Twenty years experience of surgical aortic valvotomy for critical aortic stenosis in early infancy. Eur J Cardiothorac Surg 2006;30: [11] Elder RW, Quaegebeur JM, Bacha EA, Chen JM, Bourlon F, Williams IA. Outcomes of the infant Ross procedure for congenital aortic stenosis followed into adolescence. J Thorac Cardiovasc Surg 2013;145: [12] Prijic SM, Vukomanovic VA, Stajevic MS, Bjelakovic BB, Zdravkovic MD, Sehic IN et al. Balloon dilation and surgical valvotomy comparison in noncritical congenital aortic valve stenosis. Pediatr Cardiol 2015;36: [13] Brown JW, Rodefeld MD, Ruzmetov M, Eltayeb O, Yurdakok O, Turrentine MW. Surgical valvuloplasty versus balloon aortic dilation for congenital aortic stenosis: are evidence-based outcomes relevant? Ann Thorac Surg 2012;94: [14] McCrindle BW, Blackstone EH, Williams WG, Sittiwangkul R, Spray TL, Azakie A et al. Are outcomes of surgical versus transcatheter balloon valvotomy equivalent in neonatal critical aortic stenosis? Circulation 2001; 104(Suppl 1):I [15] Haycock GB, Schwartz GJ, Wisotsky DH. Geometric method for measuring body surface area: a height-weight formula validated in infants, children, and adults. J Pediatr 1978;93:62 6. [16] Pettersen MD, Du W, Skeens ME, Humes RA. Regression equations for calculation of Z scores of cardiac structures in a large cohort of healthy infants, children, and adolescents: an echocardiographic study. J Am Soc Echocardiogr 2008;21: [17] Freedom RM (ed). The Natural and Modified History of Congenital Heart Disease. Elmsford, NY: Blackwell Pub./Futura; 2004:s140. [18] Lundström NR, Berggren H, Björkhem G, Jögi P, Sunnegårdh J. Centralization of pediatric heart surgery in Sweden. Pediatr Cardiol 2000;21: [19] Soulatges C, Momeni M, Zarrouk N, Moniotte S, Carbonez K, Barrea C et al. Long-term results of balloon valvuloplasty as primary treatment for congenital aortic valve stenosis: a 20-year review. Pediatr Cardiol 2015; 36: [20] Kallio M, Rahkonen O, Mattila I, Pihkala J. Congenital aortic stenosis: treatment outcomes in a nationwide survey. Scand Cardiovasc J 2017;51: [21] Galoin-Bertail C, Capderou A, Belli E, Houyel L. The mid-term outcome of primary open valvotomy for critical aortic stenosis in early infancy a retrospective single center study over 18 years. J Cardiothorac Surg 2016;11:116. [22] Tuo G, Khambadkone S, Tann O, Kostolny M, Derrick G, Tsang V et al. Obstructive left heart disease in neonates with a borderline left ventricle: diagnostic challenges to choosing the best outcome. Pediatr Cardiol 2013;34: [23] Lofland GK, McCrindle BW, Williams WG, Blackstone EH, Tchervenkov CI, Sittiwangkul R et al. Critical aortic stenosis in the neonate: a multiinstitutional study of management, outcomes, and risk factors. J Thorac Cardiovasc Surg 2001;121: [24] McElhinney DB, Lock JE, Keane JF, Moran AM, Colan SD. Left heart growth, function, and reintervention after balloon aortic valvuloplasty for neonatal aortic stenosis. Circulation 2005;111: [25] Petit CJ, Maskatia SA, Justino H, Mattamal RJ, Crystal MA, Ing FF. Repeat balloon aortic valvuloplasty effectively delays surgical intervention in children with recurrent aortic stenosis. Catheter Cardiovasc Interv 2013; 82: [26] Maskatia SA, Ing FF, Justino H, Crystal MA, Mullins CE, Mattamal RJ et al. Twenty-five year experience with balloon aortic valvuloplasty for congenital aortic stenosis. Am J Cardiol 2011;108: