Left Heart Growth, Function, and Reintervention After Balloon Aortic Valvuloplasty for Neonatal Aortic Stenosis
http://www.100md.com
循环学杂志 2005年第2期
the Department of Cardiology, Children’s Hospital
Department of Pediatrics, Harvard Medical School, Boston, Mass.
Abstract
Background— Transcatheter balloon aortic valvuloplasty (BAVP) has become the first-line treatment for critical aortic stenosis (AS) in neonates. However, little is known about the growth and function of left heart structures or about patterns of reintervention on the left heart after neonatal BAVP.
Methods and Results— Between 1985 and 2002, 113 patients underwent neonatal BAVP at 60 days of age. There were 16 early deaths (14%), with a significant decrease from 1985 to 1993 (22%) to 1994 to 2002 (4%), and 6 patients had successful early conversion to a univentricular circulation. In the short term, the mean relative gradient reduction was 54±26%, and significant aortic regurgitation (AR) developed in 15% of patients. The 91 early survivors with a biventricular circulation were followed up for 6.3±5.3 years, during which time there was a steady increase in the frequency of significant AR. Freedom from moderate or severe AR was 65% at 5 years. In almost all patients with a baseline aortic annulus z score less than –1, the annulus diameter increased to within the normal range within 1 to 2 years. Similarly, left ventricular (LV) end-diastolic dimension z scores, which ranged from –5 to 7.5 before BAVP, normalized within 1 to 2 years in nearly all patients with a predilation z score less than –1. Among early survivors with a biventricular circulation, reintervention-free survival on the LV outflow tract was 65% at 1 year and 48% at 5 years, with younger age, higher pre- and post-BAVP gradients, and a larger balloon-annulus diameter ratio associated with decreased reintervention-free survival (P<0.01). Seventeen surgical interventions were performed on the aortic valve in 15 patients, including replacement in 7. Survival free from aortic valve replacement was 84% at 5 years.
Conclusions— BAVP for AS during the first 60 days of life results in short-term relief of AS in the majority of patients. Among early survivors, initially small left heart structures may be associated with worse subacute outcomes but typically normalize within 1 year. Reintervention for residual/recurrent AS or iatrogenic AR is relatively common, particularly during the first year after BAVP, but aortic valve replacement during early childhood is seldom necessary.
Key Words: aorta ; stenosis ; regurgitation ; valvuloplasty
Introduction
Since its introduction almost 20 years ago, transcatheter balloon aortic valvuloplasty (BAVP) has become the first-line treatment for critical aortic valve stenosis (AS) in neonates.1–4 BAVP is comparable to surgical aortic valvuloplasty with respect to the primary outcomes of survival, efficacy at relieving AS, and frequency of important complications such as aortic regurgitation (AR).1,5 Outcomes among neonates with critical AS depend on the anatomic adequacy and functional integrity of left heart structures, as well as the effectiveness of the primary intervention.2,3,6–8 In many patients with critical AS, not only is the aortic valve stenotic and dysplastic, but the aortic valve annulus and additional components of the left heart complex are hypoplastic as well.2,3 The ability of the small left heart to support the systemic circulation effectively in both the short and long term may depend in part on the growth of left heart structures to normal dimensions. However, even if normalization of left heart dimensions occurs after BAVP, persistent or acquired functional abnormalities of the aortic valve, left ventricle (LV), and mitral valve (MV) may have an impact on the long-term outcome of these patients.
Patterns of growth of the aortic valve annulus, aortic root, and LV after neonatal BAVP have not been characterized in a large series of patients nor have the LV functional parameters or the effects of abnormal ventricular work on left heart growth or function. Similarly, there are limited data concerning the frequency and predictors of reintervention/intervention on left heart structures after neonatal BAVP. In this study, we investigated the growth and function of the left heart after BAVP for valvar AS in neonates and infants 60 days of age by assessing (1) short- and intermediate-term growth/change in dimensions of left heart structures, (2) effects of persistent AS or acquired AR on short- and intermediate-term left heart dimensions, (3) development or persistence of left heart functional abnormalities, and (4) frequency of and factors associated with reintervention on the aortic valve and/or intervention on other left heart structures.
Methods
Patients
Patients with a diagnosis of valvar AS who underwent BAVP at 60 days of age between 1985 and 2002 were ascertained from the computer database of the Department of Cardiology at Children’s Hospital, Boston, Mass. Patients with associated coarctation of the aorta were included, but those with subvalvar or supravalvar LV outflow tract obstruction, a ventricular septal defect, or other significant cardiovascular anomalies (eg, valvar atresia, abnormally related great arteries, interrupted aortic arch, total anomalous pulmonary venous return, etc) were not. Patients who had undergone prior transcatheter or surgical interventions on the aortic valve before referral to our institution were also excluded. Indications for neonatal therapy for AS included echocardiographic evidence of LV dysfunction and/or clinical symptoms of congestive heart failure. During the study period, anatomic/functional criteria for BAVP versus univentricular palliation in patients with AS and small left heart structures were not necessarily consistent at our center. However, after 1991, our published predictive equation for anatomic features associated with successful biventricular repair in neonates with AS was used as a guideline.3 Cross-sectional follow-up was obtained by December 2003.
Dimensions of Left Heart Structures
Preintervention and follow-up echocardiograms and echocardiographic reports were reviewed, and dimensions of left heart structures were recorded, according to previously reported methods.3 Measurements were indexed to body surface area, calculated according to the formula of Haycock,9 and z scores (number of standard deviations above or below the normal population mean of the structure in question indexed to body surface area) were calculated on the basis of data obtained at our institution from children with structurally and functionally normal hearts. For this study, valve morphology was not assessed. The extent of LV endocardial fibroelastosis was graded according to the system of the Congenital Heart Surgeons’ Society (CHSS).2
Functional Data
AS severity measured by echocardiography is expressed in terms of the maximum instantaneous Doppler-derived gradient, and pressure gradients measured at catheterization are expressed as peak-to-peak gradients. AR was graded according to a composite assessment scale as none-trivial, mild (no LV dilation, no retrograde flow in the descending aorta, and proximal jet width <2.5 mm/m2), moderate (LV end-diastolic volume z score >2 but <4, with or without retrograde flow in the descending aorta, and proximal jet width >2.5 but <3.5 mm/m2), or severe (LV end-diastolic volume z score >4, retrograde flow in the descending aorta, and proximal jet width >3.5 mm/m2), with emphasis placed on jet width if criteria were inconsistent. For the purposes of analysis, AR severity was collapsed into a dichotomous categorical variable: moderate or severe and less than moderate.
Balloon Aortic Valvuloplasty
The technical details of BAVP have been described previously.10–13
Data Analysis
Outcome measures included survival free from cardiovascular reintervention/intervention, survival free from reintervention on the LV outflow tract (aortic valve or subvalvar region), survival free from specific reinterventions on the aortic valve (repeated BAVP, surgical aortic valvuloplasty, aortic valve replacement), survival free from intervention on the ascending aorta/aortic arch, survival free from intervention on the MV, and development of moderate or severe AR. Follow-up estimates of reintervention-free survival were performed according to the method of Kaplan and Meier and included only patients surviving >30 days after BAVP with a biventricular circulation. Analysis of functional outcomes (ie, acute AR, residual AS, and freedom from development of moderate or severe AR) was performed for the entire cohort of patients undergoing BAVP at 60 days of age. For Kaplan-Meier analysis of freedom from moderate or severe AR, patients were censored event-free at the time of death or conversion to a univentricular circulation if they had not developed moderate or severe AR. Factors associated with time-dependent outcomes were analyzed by Cox proportional-hazards regression. In patients who underwent surgical intervention on the aortic valve, the last preoperative gradient was used as the most recent follow-up gradient for statistical analysis. Independent variables tested for association with outcome measures included demographic factors (age, weight, and treatment era as 1985 to 1993 vs 1994 to 2002), anatomic features (z scores of left heart structures, extent of endocardial fibroelastosis, associated coarctation of the aorta, or patent ductus arteriosus), functional parameters (AS gradient before BAVP or significant AR), procedural factors (antegrade vs retrograde approach, balloon-aortic annulus ratio, and number of balloon sizes and inflations) and those with respect to subacute outcomes and early functional results (residual AS gradient or acute AR). Independent variables collected as continuous data were typically analyzed as both continuous variables and collapsed into dichotomous categorical variables. Unless otherwise specified, data are presented as mean±SD or median (range).
Results
Patients
Between 1985 and 2002, 113 patients underwent BAVP for critical AS at 60 days of age, including 94 (83%) 30 days of age, 63 (56%) 1 week of age, and 73 (65%) with a patent ductus arteriosus. During the study period, 1 other patient with critical AS underwent surgical aortic valvotomy, and 3 died before intended BAVP.
Eleven patients underwent conversion to a functionally univentricular circulation 30 days after BAVP, 5 of whom died in the early postoperative period, and 11 patients died 30 days after BAVP without attempted conversion to a univentricular circulation. Thus, there were 16 early deaths (14%) and 6 patients who survived after conversion to a univentricular circulation. There were 2 early deaths (4%) among the 49 patients who underwent BAVP from 1994 to 2002, significantly fewer than among the 64 patients treated between 1985 and 1993 (n=14, 22%; P=0.007). At cross-sectional follow-up of 6.3±5.3 years, 11 additional patients had died, for a Kaplan-Meier survival of 88±3% at 1 year and 87±4% at 5 and 10 years among the 91 patients who survived the early post-BAVP period with a biventricular circulation. Among early survivors, factors independently associated with decreased survival over time included a smaller LV long-axis–heart long-axis dimension ratio (P=0.02) and a smaller aortic annulus z score (P=0.04).
Dimensions of Left Heart Structures Before BAVP
Technical Details of BAVP
Among the entire cohort of 113 patients, BAVP was performed via an antegrade approach in 35 patients (31%), a retrograde approach in 76 (67%), and a combination of antegrade and retrograde approaches in 2 (2%). The median number of balloon sizes used per patient was 2 (range, 1 to 5), and the median number of balloon inflations was 2 (range, 1 to 8). The median initial balloon-annulus diameter ratio was 0.8 (range, 0.4 to 1.0), and the median largest balloon-annulus diameter ratio was 0.9 (range, 0.7 to 1.2).
Dimensions of Left Heart Structures After BAVP
In 70 of the 91 early survivors (70%), at least 2 (6.2±4.1) serial echocardiograms beyond 1 month after BAVP were available for review and contained adequate images for measurement of the aortic annulus, aortic root, and/or LV EDD.
Aortic Dimensions
Aortic valve annulus and aortic root z scores, which varied widely before BAVP, were generally above the population mean throughout follow-up (Figure 1). There was no apparent association between aortic annulus or root z scores and post-BAVP residual AS or new-onset AR. Among the 34 patients with an aortic annulus z score less than –1 before BAVP who survived and had serial echocardiographic data available, annulus diameter typically normalized within 1 year (Figure 1).
LV Dimensions
LV EDD z scores, which ranged from –4.9 to 7.5 before BAVP, were typically above normal or in the upper stratum of the normal range during follow-up (Figure 1). Particularly notable was the normalization of LV EDD within 1 to 2 years in nearly all patients with a pre-BAVP LV EDD z score less than –1 (Figure 1). Patients with postdilation AR had a more rapid increase in LV EDD z score and tended to have higher LV EDD z scores throughout follow-up as well (Figure 1).
Functional Parameters of Left Heart Structures After BAVP
Residual/Recurrent AS After BAVP
Among the entire cohort of neonates undergoing BAVP (data available for 108 of the 113 patients), the peak gradient across the aortic valve was reduced from 57±22 to 24±12 mm Hg, P<0.001). The mean relative gradient reduction was 54±26%, and a reduction of at least 50% was achieved in 68% of patients. The early-postdilation, Doppler-derived maximum gradient was 36±14 mm Hg, and this was 13±13 mm Hg higher than the directly measured postdilation gradient (P<0.001). Factors associated with failure to achieve a gradient reduction of at least 50% included lower predilation AS gradient (47±18 vs 64±17 mm Hg, P<0.001), smaller predilation aortic annulus z score (–1.8±1.2 vs –1.0±1.3, P=0.008), and smaller predilation aortic root z score (–0.9±1.0 vs –0.3±1.1, P=0.02).
At the most recent follow-up in the 91 early survivors, the Doppler-derived maximum instantaneous gradient was 44±17 mm Hg (P=0.01 for paired t test comparison with the early postdilation, Doppler-derived gradient), not controlling for repeated BAVP.
AR After BAVP
Moderate or severe AR developed shortly after BAVP in 17 of 113 patients (15%), including 11 of 64 patients (17%) treated from 1985 to 1993 and 6 of 49 (12%) treated from 1994 to 2002 (P=0.49). Four of these patients were among the 16 early deaths. None of the patient-related or procedural variables examined were significantly associated with increased risk of short-term moderate or severe AR. Progression to moderate or severe AR occurred steadily during the follow-up period, and Kaplan-Meier freedom from moderate or severe AR was 85±3% at 1 month, 77±4% at 1 year, 65±6% at 5 years, and 61±6% at 10 years (Figure 2). The Kaplan-Meier analysis is not controlled for repeat BAVP, but only 3 of 37 patients developed acute AR after repeat BAVP, and freedom from moderate or severe AR differs minimally if these patients are censored event-free at the time of repeat BAVP. The only patient-related or procedural factor associated with decreased freedom from moderate or severe AR was larger balloon-annulus ratio at the time of BAVP.
Reinterventions/Interventions After BAVP
During a follow-up of 6.3±5.3 years, 59 of the 91 patients surviving 30 days with a biventricular circulation underwent additional cardiovascular interventions, including 58 who underwent reintervention/intervention on a left heart structure (MV, subvalvar LV outflow tract, aortic valve, ascending aorta, and/or aortic arch) and 22 who underwent multiple left heart interventions (Table 2). Survival free from any cardiovascular reintervention/intervention was 58±5% at 1 year, 38±6% at 5 years, and 26±6% at 10 years after BAVP.
Reinterventions/Interventions on the LV Outflow Tract
During follow-up, 68 reinterventions/interventions on the LV outflow tract, including the aortic valve and subvalvar region, were performed in 49 patients (Table 2). Kaplan-Meier survival free from reintervention on the LV outflow tract was 95±2% at 1 month, 65±5% at 1 year, 48±6% at 5 years, and 29±6% at 10 years, and survival free from a second LV outflow tract reintervention was 100% at 1 month, 88±3% at 1 year, 78±5% at 5 years, and 66±7% at 10 years (Figure 3A). Factors associated with decreased survival free from reintervention on the LV outflow tract included younger age at the time of BAVP (P=0.003), higher pre- (P=0.009) and post- (P=0.001) BAVP AS gradients, smaller relative reduction in the AS gradient at the initial procedure (P<0.001), and a larger balloon-annulus diameter ratio (P=0.03).
Reinterventions on the Aortic Valve
Reinterventions on the aortic valve were performed in 48 patients, with a survival free from aortic valve reintervention of 95±2% at 1 month, 65±5% at 1 year, 48±6% at 5 years, and 29±6% at 10 years (Figure 3B). The primary indication for reintervention was residual or recurrent AS in 33 patients, AR in 9 patients, and both AS and AR in 6.
Thirty-six patients underwent a total of 47 repeated BAVP procedures. The Doppler-derived maximum gradient across the aortic valve before repeated BAVP was 71±17 mm Hg, and this was 39±23 mm Hg greater than on the early post-BAVP echocardiogram (P<0.001). Kaplan-Meier survival free from repeated BAVP was 94±2% at 1 month, 65±5% at 1 year, 58±5% at 5 years, and 46±7% at 10 years (Figure 3C). Factors associated with shorter survival free from repeated BAVP included younger age (P=0.01), higher predilation AS gradient (P=0.03), and smaller relative reduction in the AS gradient at the initial procedure (P=0.001).
Among the 36 patients who underwent repeated BAVP, 16 additional interventions on the LV outflow tract were subsequently performed in 13 patients, including a third BAVP procedure in 10, surgical aortic valvuloplasty in 1, a Ross or Ross-Konno procedure in 2, and resection of subaortic stenosis in 3.
Among the 10 patients who underwent reintervention with surgical aortic valvuloplasty, 2 required subsequent aortic valve replacement (50 and 66 months later), and 8 were free from subsequent interventions on the aortic valve at a median of 45 months (range, 4 to 96 months). The maximum AS gradient in these 8 patients ranged from 15 to 75 mm Hg (median, 45 mm Hg) on the most recent echocardiogram.
Reinterventions on the Subvalvar LV Outflow Tract
Relief of subvalvar LV outflow tract obstruction was performed in 6 patients, 5 of whom had prior or concurrent interventions on the aortic valve. In 4 cases, the intervention consisted of LV outflow tract myectomy; in 1, a modified (ie, valve-sparing) Konno procedure was performed; and the other underwent a Ross-Konno procedure.
Interventions on Other Left Heart Structures
Interventions on the ascending aorta or arch were performed in 19 patients, including 17 with aortic coarctation, 8 of whom underwent surgical repair or balloon dilation 2 weeks before or on the same day as BAVP (Table 2). Including antecedent or concurrent procedures, survival free from intervention for aortic coarctation was 89±3% at 1 month, 82±4% at 1 year, 76±4% at 5 years, and 72±5% at 10 years (Figure 3B).
Discussion
Between 1985 and 2002, 113 patients underwent BAVP for critical AS at 60 days of age, with 91 early survivors. In this report, we have detailed the growth and function of the aortic valve and LV in these patients, as well as the patterns of post-BAVP intervention on the aortic valve or other left heart structures. Although the significant decrease in early mortality during the latter half of our experience is notable, survival, particularly early survival, was not the focus of this study.
Growth of Left Heart Structures After Neonatal BAVP
One of the major unresolved questions in the management of neonates with critical AS and a small aortic annulus and/or other left heart structures is the capacity and realization of left heart growth to normal dimensions. Several studies have identified small LV inflow, cavity, and/or outflow dimensions as determinants of early survival among neonates with critical AS,1–3,6–8 but little is known about either the growth of initially small left heart structures or the relation between left heart dimensions and subacute/chronic outcomes. Although the majority of our patients had aortic annulus and root diameters in the normal range (z scores less than –2 and 2) before BAVP, >65% had an aortic annulus that was smaller than normal or in the lowest portion of the normal range (ie, z score less than –1). For the most part, aortic annulus diameter in the cohort of these patients who survived the early post-BAVP period increased to the normal range within months, and over time, aortic annulus and root z scores were generally normal or only mildly enlarged.
Similarly, LV size tended to normalize rapidly after neonatal BAVP in patients spanning the spectrum of critical AS, from those with a hypoplastic, thick-walled LV to those with a dilated, poorly contracting LV. After BAVP and during follow-up, LV EDD was typically in the normal to mildly elevated range, both in patients with pre-BAVP LV hypoplasia and in those with LV dilation, with more pronounced enlargement in patients with significant AR.
Although growth of initially small left heart structures was observed in patients who were followed up with serial echocardiograms, smaller pre-BAVP left heart dimensions were also associated with certain adverse outcomes: smaller aortic annulus and aortic root z scores were correlated with a lower likelihood of achieving an AS gradient reduction 50%, and among early survivors, a smaller aortic annulus z score and LV long-axis–heart long-axis ratio were associated with decreased Kaplan-Meier survival.
Functional Status of the Aortic Valve After Neonatal BAVP
Although a substantial reduction in AS gradient was achieved in nearly all patients after neonatal BAVP, some degree of residual or recurrent AS was essentially universal. Factors associated with failure to achieve 50% gradient reduction included smaller aortic annulus and root z scores and a lower predilation AS gradient. Previous studies have found younger age (among a cohort of children of various ages) and a smaller balloon-annulus ratio to be associated with less substantial relief of AS.15 In the present cohort, neither of these variables was associated with inadequate gradient reduction. However, in the setting of critical AS, particularly in patients with LV dysfunction or a major ductal contribution to systemic blood flow, the transaortic valve gradient may not adequately reflect the severity of AS.
Acute moderate or severe AR developed after BAVP in 15% of our patients. Subsequent progression from mild or less AR to moderate or severe AR, sometimes after repeated BAVP, occurred at a steady rate over time, with 65% freedom from moderate or severe AR at 5-year follow-up. These figures are in the range of those reported in several prior series of neonates and young infants undergoing BAVP1,16,17 and mirror the finding reported in older children of chronic progression to moderate or severe AR from lesser grades in the early post-BAVP period.13,17 Although younger age has been associated with increased risk of acute AR after BAVP overall and AR seems to be more common after neonatal BAVP than after balloon dilation of congenital AS in older patients, age was not predictive of AR in our series.13,15 Similar to the findings of other investigators15,18 and to earlier reports from our center,12 a larger balloon-annulus ratio increased the risk of significant AR in this series. Although there was no threshold balloon-annulus ratio predictive of increased risk of significant AR, our general approach in neonates is not to exceed a ratio of 0.9. In the latter half of our experience, we tended to use an antegrade approach more often, beginning with a smaller balloon-annulus ratio and gradually upsizing to a ratio of 0.9; although the frequency of significant AR has decreased during this period, the trend is not statistically significant. Others have reported antegrade BAVP to be associated with less severe acute AR,19 and although >30% of our patients underwent BAVP via an antegrade approach, there was no difference in the frequency of significant AR after antegrade and retrograde BAVP.
Reinterventions on the Left Heart After Neonatal BAVP
Reinterventions on the left heart are common among early survivors of neonatal BAVP, with the majority performed for residual/recurrent AS or AR. At 1-year follow-up, overall reintervention-free survival was 58%, and survival free from aortic valve reintervention was 65%, with the steepest hazard during the first 6 months. This early hazard was a function of 2 primary trends: (1) repeated BAVP within this time frame, which reflects a tendency to leave patients with residual AS at the initial BAVP procedure rather than risk significant AR in the pursuit of more complete relief of AS, and (2) early intervention in a subset of patients with aortic coarctation. In addition, there was a steady early hazard for surgical reintervention on the aortic valve during the first year, invariably in patients with significant post-BAVP AR. Beyond the first year, there was a steady, less acute decrease in survival free from repeated BAVP, whereas there was a relative surge in surgical reinterventions on the aortic valve between 3 and 6 years after BAVP, after which these, too, leveled off. Similar to the findings of the CHSS (13 of our patients were included in the CHSS study cohort), survival free from reintervention on the aortic valve was 50% at 5 years.1
Survival free from repeated BAVP was 65% at 1 year and 58% at 5 years. However, we do not consider this finding to reflect inadequate primary treatment. In fact, over time, as our tendency to avoid dilation with a balloon-annulus ratio >0.9 has become more entrenched, we have come to accept repeated BAVP as preferable to the potential consequences of more aggressive dilation during the initial BAVP procedure. Risk factors for shorter survival free from repeated BAVP included younger age, a higher predilation AS gradient, and a smaller relative reduction in the gradient at the initial procedure.
Freedom from surgical reintervention on the aortic valve in this study was 74% at 5 years, similar to that reported by Hawkins et al18 in a series consisting primarily of patients undergoing BAVP at an older age. Only acute postdilation moderate or severe AR was associated with shorter freedom from surgical reintervention on the aortic valve. Ten patients underwent surgical aortic valvuloplasty, typically reconstructive procedures for AR, such as leaflet augmentation and/or resuspension. As previously reported, such reconstructive techniques in patients with an abnormal aortic valve and AR after BAVP generally impart sufficient valve function to defer aortic valve replacement for at least 4 to 5 years.14 Only 2 of these 10 patients and 5 others underwent aortic valve replacement.
Aside from procedures on the aortic valve, left heart interventions were relatively uncommon. Aortic coarctation was the most frequent non–valve-related indication for intervention, with coarctation repairs concentrated in the neonatal period. Interventions for subaortic stenosis were performed in 6 patients, 5 of whom underwent concurrent intervention on the aortic valve. Interventions for mitral stenosis (n=5) or dilation-related mitral regurgitation (n=1) were performed in 6 patients, all within 1 year of BAVP. Thus, although there was a spectrum of left heart hypoplasia and left-sided anomalies in neonates with AS, in most cases these associated findings did not necessitate intervention.
Limitations of the Study
The primary limitations to this study are its retrospective design and the extended enrollment period (1985 to 2002), logistical considerations that may limit the applicability of our findings. For example, there has been substantial evolution in balloon technology and experience with interventional therapy in pediatric cardiology during the duration of this study, which may have influenced both technical variables and outcomes. Similarly, indications for BAVP versus univentricular palliation among patients with AS and small left heart structures were not necessarily consistent over the course of the study period.
Conclusions
BAVP for AS during the first 60 days of life results in acute relief of AS in the majority of patients. Among early survivors, initially small left heart structures may be associated with worse subacute outcomes, but typically, these normalize within 1 year. Reintervention for residual/recurrent AS or iatrogenic AR is relatively common, particularly during the first year after BAVP, but aortic valve replacement during early childhood is required infrequently.
Footnotes
Guest Editor for this article was Robin J. Barst, MD.
References
McCrindle BW, Blackstone EH, Williams WG, Sittiwangkul R, Spray TL, Azakie A, Jonas RA. Are outcomes of surgical versus transcatheter balloon valvotomy equivalent in neonatal critical aortic stenosis; Circulation. 2001; 104 (suppl I): I-152–I-158.
Lofland GK, McCrindle BW, Williams WG, Blackstone EH, Tchervenkov CI, Sittiwangkul R, Jonas RA. Critical aortic stenosis in the neonate: a multi-institutional study of management, outcomes, and risk factors; Congenital Heart Surgeons Society. J Thorac Cardiovasc Surg. 2001; 121: 10–27.
Rhodes LA, Colan SD, Perry SB, Jonas RA, Sanders SP. Predictors of survival in neonates with critical aortic stenosis. Circulation. 1991; 84: 2325–2335.
Bu’Lock FA, Joffe HS, Jordan SC, Martin RP. Balloon dilatation (valvoplasty) as first line treatment for severe stenosis of the aortic valve in early infancy: medium term results and determinants of survival. Br Heart J. 1993; 70: 546–553.
Mosca RS, Iannettoni MD, Schwartz SM, Ludomirsky A, Beekman RH, Lloyd T, Bove EL. Critical aortic stenosis in the neonate: a comparison of balloon valvuloplasty and transventricular dilation. J Thorac Cardiovasc Surg. 1995; 109: 147–154.
Latson LA, Cheatham JP, Gutgesell HP. Relation of the echocardiographic estimate of left ventricular size to mortality in infants with severe left ventricular outflow obstruction. Am J Cardiol. 1981; 48: 887–891.
Pelech AN, Dyck JD, Trusler GA, Williams WG, Olley PM, Rowe RD, Freedom RM. Critical aortic stenosis: survival and management. J Thorac Cardiovasc Surg. 1987; 94: 510–517.
Parsons MK, Moreau GA, Graham TP Jr, Johns JA, Boucek RJ. Echocardiographic estimation of critical left ventricular size in infants with isolated aortic valve stenosis. J Am Coll Cardiol. 1991; 18: 1049–1055.
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: 62–66.
Zeevi B, Keane JF, Castaneda AR, Perry SB, Lock JE. Neonatal critical valvar aortic stenosis: a comparison of surgical and balloon dilation therapy. Circulation. 1989; 80: 831–839.
Egito ES, Moore P, O’Sullivan J, Colan S, Perry SB, Lock JE, Keane JF. Transvascular balloon dilation for neonatal critical aortic stenosis: early and midterm results. J Am Coll Cardiol. 1997; 29: 442–447.
Sholler GF, Keane JF, Perry SB, Sanders SP, Lock JE. Balloon dilation of congenital aortic valve stenosis: results and influence of technical and morphological features on outcome. Circulation. 1988; 78: 351–360.
Moore P, Egito E, Mowrey H, Perry SB, Lock JE, Keane JF. Midterm results of balloon dilation of congenital aortic stenosis: predictors of success. J Am Coll Cardiol. 1996; 27: 1257–1263.
Bacha EA, Satou GM, Moran AM, Zurakowski D, Marx GR, Keane JF, Jonas RA. Valve-sparing operation for balloon-induced aortic regurgitation in congenital aortic stenosis. J Thorac Cardiovasc Surg. 2001; 122: 162–168.
McCrindle BW. Independent predictors of immediate results of percutaneous balloon aortic valvotomy in children; Valvuloplasty and Angioplasty of Congenital Anomalies (VACA) Registry Investigators. Am J Cardiol. 1996; 77: 286–293.
Peuster M, Fink C, Schoof S, Von Schnakenburg C, Hausdorf G. Anterograde balloon valvuloplasty for the treatment of neonatal critical valvar aortic stenosis. Catheter Cardiovasc Interv. 2002; 56: 516–520.
Balmer C, Beghetti M, Fasnacht M, Friedli B, Arbenz U. Balloon aortic valvoplasty in paediatric patients: progressive aortic regurgitation is common. Heart. 2004; 90: 77–81.
Hawkins JA, Minich LL, Shaddy RE, Tani LY, Orsmond GS, Sturtevant JE, McGough EC. Aortic valve repair and replacement after balloon aortic valvuloplasty in children. Ann Thorac Surg. 1996; 61: 1355–1358.
Magee AG, Nykanen D, McCrindle BW, Wax D, Freedom RM, Benson LN. Balloon dilation of severe aortic stenosis in the neonate: comparison of anterograde and retrograde catheter approaches. J Am Coll Cardiol. 1997; 30: 1061–1066.(Doff B. McElhinney, MD; J)
Department of Pediatrics, Harvard Medical School, Boston, Mass.
Abstract
Background— Transcatheter balloon aortic valvuloplasty (BAVP) has become the first-line treatment for critical aortic stenosis (AS) in neonates. However, little is known about the growth and function of left heart structures or about patterns of reintervention on the left heart after neonatal BAVP.
Methods and Results— Between 1985 and 2002, 113 patients underwent neonatal BAVP at 60 days of age. There were 16 early deaths (14%), with a significant decrease from 1985 to 1993 (22%) to 1994 to 2002 (4%), and 6 patients had successful early conversion to a univentricular circulation. In the short term, the mean relative gradient reduction was 54±26%, and significant aortic regurgitation (AR) developed in 15% of patients. The 91 early survivors with a biventricular circulation were followed up for 6.3±5.3 years, during which time there was a steady increase in the frequency of significant AR. Freedom from moderate or severe AR was 65% at 5 years. In almost all patients with a baseline aortic annulus z score less than –1, the annulus diameter increased to within the normal range within 1 to 2 years. Similarly, left ventricular (LV) end-diastolic dimension z scores, which ranged from –5 to 7.5 before BAVP, normalized within 1 to 2 years in nearly all patients with a predilation z score less than –1. Among early survivors with a biventricular circulation, reintervention-free survival on the LV outflow tract was 65% at 1 year and 48% at 5 years, with younger age, higher pre- and post-BAVP gradients, and a larger balloon-annulus diameter ratio associated with decreased reintervention-free survival (P<0.01). Seventeen surgical interventions were performed on the aortic valve in 15 patients, including replacement in 7. Survival free from aortic valve replacement was 84% at 5 years.
Conclusions— BAVP for AS during the first 60 days of life results in short-term relief of AS in the majority of patients. Among early survivors, initially small left heart structures may be associated with worse subacute outcomes but typically normalize within 1 year. Reintervention for residual/recurrent AS or iatrogenic AR is relatively common, particularly during the first year after BAVP, but aortic valve replacement during early childhood is seldom necessary.
Key Words: aorta ; stenosis ; regurgitation ; valvuloplasty
Introduction
Since its introduction almost 20 years ago, transcatheter balloon aortic valvuloplasty (BAVP) has become the first-line treatment for critical aortic valve stenosis (AS) in neonates.1–4 BAVP is comparable to surgical aortic valvuloplasty with respect to the primary outcomes of survival, efficacy at relieving AS, and frequency of important complications such as aortic regurgitation (AR).1,5 Outcomes among neonates with critical AS depend on the anatomic adequacy and functional integrity of left heart structures, as well as the effectiveness of the primary intervention.2,3,6–8 In many patients with critical AS, not only is the aortic valve stenotic and dysplastic, but the aortic valve annulus and additional components of the left heart complex are hypoplastic as well.2,3 The ability of the small left heart to support the systemic circulation effectively in both the short and long term may depend in part on the growth of left heart structures to normal dimensions. However, even if normalization of left heart dimensions occurs after BAVP, persistent or acquired functional abnormalities of the aortic valve, left ventricle (LV), and mitral valve (MV) may have an impact on the long-term outcome of these patients.
Patterns of growth of the aortic valve annulus, aortic root, and LV after neonatal BAVP have not been characterized in a large series of patients nor have the LV functional parameters or the effects of abnormal ventricular work on left heart growth or function. Similarly, there are limited data concerning the frequency and predictors of reintervention/intervention on left heart structures after neonatal BAVP. In this study, we investigated the growth and function of the left heart after BAVP for valvar AS in neonates and infants 60 days of age by assessing (1) short- and intermediate-term growth/change in dimensions of left heart structures, (2) effects of persistent AS or acquired AR on short- and intermediate-term left heart dimensions, (3) development or persistence of left heart functional abnormalities, and (4) frequency of and factors associated with reintervention on the aortic valve and/or intervention on other left heart structures.
Methods
Patients
Patients with a diagnosis of valvar AS who underwent BAVP at 60 days of age between 1985 and 2002 were ascertained from the computer database of the Department of Cardiology at Children’s Hospital, Boston, Mass. Patients with associated coarctation of the aorta were included, but those with subvalvar or supravalvar LV outflow tract obstruction, a ventricular septal defect, or other significant cardiovascular anomalies (eg, valvar atresia, abnormally related great arteries, interrupted aortic arch, total anomalous pulmonary venous return, etc) were not. Patients who had undergone prior transcatheter or surgical interventions on the aortic valve before referral to our institution were also excluded. Indications for neonatal therapy for AS included echocardiographic evidence of LV dysfunction and/or clinical symptoms of congestive heart failure. During the study period, anatomic/functional criteria for BAVP versus univentricular palliation in patients with AS and small left heart structures were not necessarily consistent at our center. However, after 1991, our published predictive equation for anatomic features associated with successful biventricular repair in neonates with AS was used as a guideline.3 Cross-sectional follow-up was obtained by December 2003.
Dimensions of Left Heart Structures
Preintervention and follow-up echocardiograms and echocardiographic reports were reviewed, and dimensions of left heart structures were recorded, according to previously reported methods.3 Measurements were indexed to body surface area, calculated according to the formula of Haycock,9 and z scores (number of standard deviations above or below the normal population mean of the structure in question indexed to body surface area) were calculated on the basis of data obtained at our institution from children with structurally and functionally normal hearts. For this study, valve morphology was not assessed. The extent of LV endocardial fibroelastosis was graded according to the system of the Congenital Heart Surgeons’ Society (CHSS).2
Functional Data
AS severity measured by echocardiography is expressed in terms of the maximum instantaneous Doppler-derived gradient, and pressure gradients measured at catheterization are expressed as peak-to-peak gradients. AR was graded according to a composite assessment scale as none-trivial, mild (no LV dilation, no retrograde flow in the descending aorta, and proximal jet width <2.5 mm/m2), moderate (LV end-diastolic volume z score >2 but <4, with or without retrograde flow in the descending aorta, and proximal jet width >2.5 but <3.5 mm/m2), or severe (LV end-diastolic volume z score >4, retrograde flow in the descending aorta, and proximal jet width >3.5 mm/m2), with emphasis placed on jet width if criteria were inconsistent. For the purposes of analysis, AR severity was collapsed into a dichotomous categorical variable: moderate or severe and less than moderate.
Balloon Aortic Valvuloplasty
The technical details of BAVP have been described previously.10–13
Data Analysis
Outcome measures included survival free from cardiovascular reintervention/intervention, survival free from reintervention on the LV outflow tract (aortic valve or subvalvar region), survival free from specific reinterventions on the aortic valve (repeated BAVP, surgical aortic valvuloplasty, aortic valve replacement), survival free from intervention on the ascending aorta/aortic arch, survival free from intervention on the MV, and development of moderate or severe AR. Follow-up estimates of reintervention-free survival were performed according to the method of Kaplan and Meier and included only patients surviving >30 days after BAVP with a biventricular circulation. Analysis of functional outcomes (ie, acute AR, residual AS, and freedom from development of moderate or severe AR) was performed for the entire cohort of patients undergoing BAVP at 60 days of age. For Kaplan-Meier analysis of freedom from moderate or severe AR, patients were censored event-free at the time of death or conversion to a univentricular circulation if they had not developed moderate or severe AR. Factors associated with time-dependent outcomes were analyzed by Cox proportional-hazards regression. In patients who underwent surgical intervention on the aortic valve, the last preoperative gradient was used as the most recent follow-up gradient for statistical analysis. Independent variables tested for association with outcome measures included demographic factors (age, weight, and treatment era as 1985 to 1993 vs 1994 to 2002), anatomic features (z scores of left heart structures, extent of endocardial fibroelastosis, associated coarctation of the aorta, or patent ductus arteriosus), functional parameters (AS gradient before BAVP or significant AR), procedural factors (antegrade vs retrograde approach, balloon-aortic annulus ratio, and number of balloon sizes and inflations) and those with respect to subacute outcomes and early functional results (residual AS gradient or acute AR). Independent variables collected as continuous data were typically analyzed as both continuous variables and collapsed into dichotomous categorical variables. Unless otherwise specified, data are presented as mean±SD or median (range).
Results
Patients
Between 1985 and 2002, 113 patients underwent BAVP for critical AS at 60 days of age, including 94 (83%) 30 days of age, 63 (56%) 1 week of age, and 73 (65%) with a patent ductus arteriosus. During the study period, 1 other patient with critical AS underwent surgical aortic valvotomy, and 3 died before intended BAVP.
Eleven patients underwent conversion to a functionally univentricular circulation 30 days after BAVP, 5 of whom died in the early postoperative period, and 11 patients died 30 days after BAVP without attempted conversion to a univentricular circulation. Thus, there were 16 early deaths (14%) and 6 patients who survived after conversion to a univentricular circulation. There were 2 early deaths (4%) among the 49 patients who underwent BAVP from 1994 to 2002, significantly fewer than among the 64 patients treated between 1985 and 1993 (n=14, 22%; P=0.007). At cross-sectional follow-up of 6.3±5.3 years, 11 additional patients had died, for a Kaplan-Meier survival of 88±3% at 1 year and 87±4% at 5 and 10 years among the 91 patients who survived the early post-BAVP period with a biventricular circulation. Among early survivors, factors independently associated with decreased survival over time included a smaller LV long-axis–heart long-axis dimension ratio (P=0.02) and a smaller aortic annulus z score (P=0.04).
Dimensions of Left Heart Structures Before BAVP
Technical Details of BAVP
Among the entire cohort of 113 patients, BAVP was performed via an antegrade approach in 35 patients (31%), a retrograde approach in 76 (67%), and a combination of antegrade and retrograde approaches in 2 (2%). The median number of balloon sizes used per patient was 2 (range, 1 to 5), and the median number of balloon inflations was 2 (range, 1 to 8). The median initial balloon-annulus diameter ratio was 0.8 (range, 0.4 to 1.0), and the median largest balloon-annulus diameter ratio was 0.9 (range, 0.7 to 1.2).
Dimensions of Left Heart Structures After BAVP
In 70 of the 91 early survivors (70%), at least 2 (6.2±4.1) serial echocardiograms beyond 1 month after BAVP were available for review and contained adequate images for measurement of the aortic annulus, aortic root, and/or LV EDD.
Aortic Dimensions
Aortic valve annulus and aortic root z scores, which varied widely before BAVP, were generally above the population mean throughout follow-up (Figure 1). There was no apparent association between aortic annulus or root z scores and post-BAVP residual AS or new-onset AR. Among the 34 patients with an aortic annulus z score less than –1 before BAVP who survived and had serial echocardiographic data available, annulus diameter typically normalized within 1 year (Figure 1).
LV Dimensions
LV EDD z scores, which ranged from –4.9 to 7.5 before BAVP, were typically above normal or in the upper stratum of the normal range during follow-up (Figure 1). Particularly notable was the normalization of LV EDD within 1 to 2 years in nearly all patients with a pre-BAVP LV EDD z score less than –1 (Figure 1). Patients with postdilation AR had a more rapid increase in LV EDD z score and tended to have higher LV EDD z scores throughout follow-up as well (Figure 1).
Functional Parameters of Left Heart Structures After BAVP
Residual/Recurrent AS After BAVP
Among the entire cohort of neonates undergoing BAVP (data available for 108 of the 113 patients), the peak gradient across the aortic valve was reduced from 57±22 to 24±12 mm Hg, P<0.001). The mean relative gradient reduction was 54±26%, and a reduction of at least 50% was achieved in 68% of patients. The early-postdilation, Doppler-derived maximum gradient was 36±14 mm Hg, and this was 13±13 mm Hg higher than the directly measured postdilation gradient (P<0.001). Factors associated with failure to achieve a gradient reduction of at least 50% included lower predilation AS gradient (47±18 vs 64±17 mm Hg, P<0.001), smaller predilation aortic annulus z score (–1.8±1.2 vs –1.0±1.3, P=0.008), and smaller predilation aortic root z score (–0.9±1.0 vs –0.3±1.1, P=0.02).
At the most recent follow-up in the 91 early survivors, the Doppler-derived maximum instantaneous gradient was 44±17 mm Hg (P=0.01 for paired t test comparison with the early postdilation, Doppler-derived gradient), not controlling for repeated BAVP.
AR After BAVP
Moderate or severe AR developed shortly after BAVP in 17 of 113 patients (15%), including 11 of 64 patients (17%) treated from 1985 to 1993 and 6 of 49 (12%) treated from 1994 to 2002 (P=0.49). Four of these patients were among the 16 early deaths. None of the patient-related or procedural variables examined were significantly associated with increased risk of short-term moderate or severe AR. Progression to moderate or severe AR occurred steadily during the follow-up period, and Kaplan-Meier freedom from moderate or severe AR was 85±3% at 1 month, 77±4% at 1 year, 65±6% at 5 years, and 61±6% at 10 years (Figure 2). The Kaplan-Meier analysis is not controlled for repeat BAVP, but only 3 of 37 patients developed acute AR after repeat BAVP, and freedom from moderate or severe AR differs minimally if these patients are censored event-free at the time of repeat BAVP. The only patient-related or procedural factor associated with decreased freedom from moderate or severe AR was larger balloon-annulus ratio at the time of BAVP.
Reinterventions/Interventions After BAVP
During a follow-up of 6.3±5.3 years, 59 of the 91 patients surviving 30 days with a biventricular circulation underwent additional cardiovascular interventions, including 58 who underwent reintervention/intervention on a left heart structure (MV, subvalvar LV outflow tract, aortic valve, ascending aorta, and/or aortic arch) and 22 who underwent multiple left heart interventions (Table 2). Survival free from any cardiovascular reintervention/intervention was 58±5% at 1 year, 38±6% at 5 years, and 26±6% at 10 years after BAVP.
Reinterventions/Interventions on the LV Outflow Tract
During follow-up, 68 reinterventions/interventions on the LV outflow tract, including the aortic valve and subvalvar region, were performed in 49 patients (Table 2). Kaplan-Meier survival free from reintervention on the LV outflow tract was 95±2% at 1 month, 65±5% at 1 year, 48±6% at 5 years, and 29±6% at 10 years, and survival free from a second LV outflow tract reintervention was 100% at 1 month, 88±3% at 1 year, 78±5% at 5 years, and 66±7% at 10 years (Figure 3A). Factors associated with decreased survival free from reintervention on the LV outflow tract included younger age at the time of BAVP (P=0.003), higher pre- (P=0.009) and post- (P=0.001) BAVP AS gradients, smaller relative reduction in the AS gradient at the initial procedure (P<0.001), and a larger balloon-annulus diameter ratio (P=0.03).
Reinterventions on the Aortic Valve
Reinterventions on the aortic valve were performed in 48 patients, with a survival free from aortic valve reintervention of 95±2% at 1 month, 65±5% at 1 year, 48±6% at 5 years, and 29±6% at 10 years (Figure 3B). The primary indication for reintervention was residual or recurrent AS in 33 patients, AR in 9 patients, and both AS and AR in 6.
Thirty-six patients underwent a total of 47 repeated BAVP procedures. The Doppler-derived maximum gradient across the aortic valve before repeated BAVP was 71±17 mm Hg, and this was 39±23 mm Hg greater than on the early post-BAVP echocardiogram (P<0.001). Kaplan-Meier survival free from repeated BAVP was 94±2% at 1 month, 65±5% at 1 year, 58±5% at 5 years, and 46±7% at 10 years (Figure 3C). Factors associated with shorter survival free from repeated BAVP included younger age (P=0.01), higher predilation AS gradient (P=0.03), and smaller relative reduction in the AS gradient at the initial procedure (P=0.001).
Among the 36 patients who underwent repeated BAVP, 16 additional interventions on the LV outflow tract were subsequently performed in 13 patients, including a third BAVP procedure in 10, surgical aortic valvuloplasty in 1, a Ross or Ross-Konno procedure in 2, and resection of subaortic stenosis in 3.
Among the 10 patients who underwent reintervention with surgical aortic valvuloplasty, 2 required subsequent aortic valve replacement (50 and 66 months later), and 8 were free from subsequent interventions on the aortic valve at a median of 45 months (range, 4 to 96 months). The maximum AS gradient in these 8 patients ranged from 15 to 75 mm Hg (median, 45 mm Hg) on the most recent echocardiogram.
Reinterventions on the Subvalvar LV Outflow Tract
Relief of subvalvar LV outflow tract obstruction was performed in 6 patients, 5 of whom had prior or concurrent interventions on the aortic valve. In 4 cases, the intervention consisted of LV outflow tract myectomy; in 1, a modified (ie, valve-sparing) Konno procedure was performed; and the other underwent a Ross-Konno procedure.
Interventions on Other Left Heart Structures
Interventions on the ascending aorta or arch were performed in 19 patients, including 17 with aortic coarctation, 8 of whom underwent surgical repair or balloon dilation 2 weeks before or on the same day as BAVP (Table 2). Including antecedent or concurrent procedures, survival free from intervention for aortic coarctation was 89±3% at 1 month, 82±4% at 1 year, 76±4% at 5 years, and 72±5% at 10 years (Figure 3B).
Discussion
Between 1985 and 2002, 113 patients underwent BAVP for critical AS at 60 days of age, with 91 early survivors. In this report, we have detailed the growth and function of the aortic valve and LV in these patients, as well as the patterns of post-BAVP intervention on the aortic valve or other left heart structures. Although the significant decrease in early mortality during the latter half of our experience is notable, survival, particularly early survival, was not the focus of this study.
Growth of Left Heart Structures After Neonatal BAVP
One of the major unresolved questions in the management of neonates with critical AS and a small aortic annulus and/or other left heart structures is the capacity and realization of left heart growth to normal dimensions. Several studies have identified small LV inflow, cavity, and/or outflow dimensions as determinants of early survival among neonates with critical AS,1–3,6–8 but little is known about either the growth of initially small left heart structures or the relation between left heart dimensions and subacute/chronic outcomes. Although the majority of our patients had aortic annulus and root diameters in the normal range (z scores less than –2 and 2) before BAVP, >65% had an aortic annulus that was smaller than normal or in the lowest portion of the normal range (ie, z score less than –1). For the most part, aortic annulus diameter in the cohort of these patients who survived the early post-BAVP period increased to the normal range within months, and over time, aortic annulus and root z scores were generally normal or only mildly enlarged.
Similarly, LV size tended to normalize rapidly after neonatal BAVP in patients spanning the spectrum of critical AS, from those with a hypoplastic, thick-walled LV to those with a dilated, poorly contracting LV. After BAVP and during follow-up, LV EDD was typically in the normal to mildly elevated range, both in patients with pre-BAVP LV hypoplasia and in those with LV dilation, with more pronounced enlargement in patients with significant AR.
Although growth of initially small left heart structures was observed in patients who were followed up with serial echocardiograms, smaller pre-BAVP left heart dimensions were also associated with certain adverse outcomes: smaller aortic annulus and aortic root z scores were correlated with a lower likelihood of achieving an AS gradient reduction 50%, and among early survivors, a smaller aortic annulus z score and LV long-axis–heart long-axis ratio were associated with decreased Kaplan-Meier survival.
Functional Status of the Aortic Valve After Neonatal BAVP
Although a substantial reduction in AS gradient was achieved in nearly all patients after neonatal BAVP, some degree of residual or recurrent AS was essentially universal. Factors associated with failure to achieve 50% gradient reduction included smaller aortic annulus and root z scores and a lower predilation AS gradient. Previous studies have found younger age (among a cohort of children of various ages) and a smaller balloon-annulus ratio to be associated with less substantial relief of AS.15 In the present cohort, neither of these variables was associated with inadequate gradient reduction. However, in the setting of critical AS, particularly in patients with LV dysfunction or a major ductal contribution to systemic blood flow, the transaortic valve gradient may not adequately reflect the severity of AS.
Acute moderate or severe AR developed after BAVP in 15% of our patients. Subsequent progression from mild or less AR to moderate or severe AR, sometimes after repeated BAVP, occurred at a steady rate over time, with 65% freedom from moderate or severe AR at 5-year follow-up. These figures are in the range of those reported in several prior series of neonates and young infants undergoing BAVP1,16,17 and mirror the finding reported in older children of chronic progression to moderate or severe AR from lesser grades in the early post-BAVP period.13,17 Although younger age has been associated with increased risk of acute AR after BAVP overall and AR seems to be more common after neonatal BAVP than after balloon dilation of congenital AS in older patients, age was not predictive of AR in our series.13,15 Similar to the findings of other investigators15,18 and to earlier reports from our center,12 a larger balloon-annulus ratio increased the risk of significant AR in this series. Although there was no threshold balloon-annulus ratio predictive of increased risk of significant AR, our general approach in neonates is not to exceed a ratio of 0.9. In the latter half of our experience, we tended to use an antegrade approach more often, beginning with a smaller balloon-annulus ratio and gradually upsizing to a ratio of 0.9; although the frequency of significant AR has decreased during this period, the trend is not statistically significant. Others have reported antegrade BAVP to be associated with less severe acute AR,19 and although >30% of our patients underwent BAVP via an antegrade approach, there was no difference in the frequency of significant AR after antegrade and retrograde BAVP.
Reinterventions on the Left Heart After Neonatal BAVP
Reinterventions on the left heart are common among early survivors of neonatal BAVP, with the majority performed for residual/recurrent AS or AR. At 1-year follow-up, overall reintervention-free survival was 58%, and survival free from aortic valve reintervention was 65%, with the steepest hazard during the first 6 months. This early hazard was a function of 2 primary trends: (1) repeated BAVP within this time frame, which reflects a tendency to leave patients with residual AS at the initial BAVP procedure rather than risk significant AR in the pursuit of more complete relief of AS, and (2) early intervention in a subset of patients with aortic coarctation. In addition, there was a steady early hazard for surgical reintervention on the aortic valve during the first year, invariably in patients with significant post-BAVP AR. Beyond the first year, there was a steady, less acute decrease in survival free from repeated BAVP, whereas there was a relative surge in surgical reinterventions on the aortic valve between 3 and 6 years after BAVP, after which these, too, leveled off. Similar to the findings of the CHSS (13 of our patients were included in the CHSS study cohort), survival free from reintervention on the aortic valve was 50% at 5 years.1
Survival free from repeated BAVP was 65% at 1 year and 58% at 5 years. However, we do not consider this finding to reflect inadequate primary treatment. In fact, over time, as our tendency to avoid dilation with a balloon-annulus ratio >0.9 has become more entrenched, we have come to accept repeated BAVP as preferable to the potential consequences of more aggressive dilation during the initial BAVP procedure. Risk factors for shorter survival free from repeated BAVP included younger age, a higher predilation AS gradient, and a smaller relative reduction in the gradient at the initial procedure.
Freedom from surgical reintervention on the aortic valve in this study was 74% at 5 years, similar to that reported by Hawkins et al18 in a series consisting primarily of patients undergoing BAVP at an older age. Only acute postdilation moderate or severe AR was associated with shorter freedom from surgical reintervention on the aortic valve. Ten patients underwent surgical aortic valvuloplasty, typically reconstructive procedures for AR, such as leaflet augmentation and/or resuspension. As previously reported, such reconstructive techniques in patients with an abnormal aortic valve and AR after BAVP generally impart sufficient valve function to defer aortic valve replacement for at least 4 to 5 years.14 Only 2 of these 10 patients and 5 others underwent aortic valve replacement.
Aside from procedures on the aortic valve, left heart interventions were relatively uncommon. Aortic coarctation was the most frequent non–valve-related indication for intervention, with coarctation repairs concentrated in the neonatal period. Interventions for subaortic stenosis were performed in 6 patients, 5 of whom underwent concurrent intervention on the aortic valve. Interventions for mitral stenosis (n=5) or dilation-related mitral regurgitation (n=1) were performed in 6 patients, all within 1 year of BAVP. Thus, although there was a spectrum of left heart hypoplasia and left-sided anomalies in neonates with AS, in most cases these associated findings did not necessitate intervention.
Limitations of the Study
The primary limitations to this study are its retrospective design and the extended enrollment period (1985 to 2002), logistical considerations that may limit the applicability of our findings. For example, there has been substantial evolution in balloon technology and experience with interventional therapy in pediatric cardiology during the duration of this study, which may have influenced both technical variables and outcomes. Similarly, indications for BAVP versus univentricular palliation among patients with AS and small left heart structures were not necessarily consistent over the course of the study period.
Conclusions
BAVP for AS during the first 60 days of life results in acute relief of AS in the majority of patients. Among early survivors, initially small left heart structures may be associated with worse subacute outcomes, but typically, these normalize within 1 year. Reintervention for residual/recurrent AS or iatrogenic AR is relatively common, particularly during the first year after BAVP, but aortic valve replacement during early childhood is required infrequently.
Footnotes
Guest Editor for this article was Robin J. Barst, MD.
References
McCrindle BW, Blackstone EH, Williams WG, Sittiwangkul R, Spray TL, Azakie A, Jonas RA. Are outcomes of surgical versus transcatheter balloon valvotomy equivalent in neonatal critical aortic stenosis; Circulation. 2001; 104 (suppl I): I-152–I-158.
Lofland GK, McCrindle BW, Williams WG, Blackstone EH, Tchervenkov CI, Sittiwangkul R, Jonas RA. Critical aortic stenosis in the neonate: a multi-institutional study of management, outcomes, and risk factors; Congenital Heart Surgeons Society. J Thorac Cardiovasc Surg. 2001; 121: 10–27.
Rhodes LA, Colan SD, Perry SB, Jonas RA, Sanders SP. Predictors of survival in neonates with critical aortic stenosis. Circulation. 1991; 84: 2325–2335.
Bu’Lock FA, Joffe HS, Jordan SC, Martin RP. Balloon dilatation (valvoplasty) as first line treatment for severe stenosis of the aortic valve in early infancy: medium term results and determinants of survival. Br Heart J. 1993; 70: 546–553.
Mosca RS, Iannettoni MD, Schwartz SM, Ludomirsky A, Beekman RH, Lloyd T, Bove EL. Critical aortic stenosis in the neonate: a comparison of balloon valvuloplasty and transventricular dilation. J Thorac Cardiovasc Surg. 1995; 109: 147–154.
Latson LA, Cheatham JP, Gutgesell HP. Relation of the echocardiographic estimate of left ventricular size to mortality in infants with severe left ventricular outflow obstruction. Am J Cardiol. 1981; 48: 887–891.
Pelech AN, Dyck JD, Trusler GA, Williams WG, Olley PM, Rowe RD, Freedom RM. Critical aortic stenosis: survival and management. J Thorac Cardiovasc Surg. 1987; 94: 510–517.
Parsons MK, Moreau GA, Graham TP Jr, Johns JA, Boucek RJ. Echocardiographic estimation of critical left ventricular size in infants with isolated aortic valve stenosis. J Am Coll Cardiol. 1991; 18: 1049–1055.
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: 62–66.
Zeevi B, Keane JF, Castaneda AR, Perry SB, Lock JE. Neonatal critical valvar aortic stenosis: a comparison of surgical and balloon dilation therapy. Circulation. 1989; 80: 831–839.
Egito ES, Moore P, O’Sullivan J, Colan S, Perry SB, Lock JE, Keane JF. Transvascular balloon dilation for neonatal critical aortic stenosis: early and midterm results. J Am Coll Cardiol. 1997; 29: 442–447.
Sholler GF, Keane JF, Perry SB, Sanders SP, Lock JE. Balloon dilation of congenital aortic valve stenosis: results and influence of technical and morphological features on outcome. Circulation. 1988; 78: 351–360.
Moore P, Egito E, Mowrey H, Perry SB, Lock JE, Keane JF. Midterm results of balloon dilation of congenital aortic stenosis: predictors of success. J Am Coll Cardiol. 1996; 27: 1257–1263.
Bacha EA, Satou GM, Moran AM, Zurakowski D, Marx GR, Keane JF, Jonas RA. Valve-sparing operation for balloon-induced aortic regurgitation in congenital aortic stenosis. J Thorac Cardiovasc Surg. 2001; 122: 162–168.
McCrindle BW. Independent predictors of immediate results of percutaneous balloon aortic valvotomy in children; Valvuloplasty and Angioplasty of Congenital Anomalies (VACA) Registry Investigators. Am J Cardiol. 1996; 77: 286–293.
Peuster M, Fink C, Schoof S, Von Schnakenburg C, Hausdorf G. Anterograde balloon valvuloplasty for the treatment of neonatal critical valvar aortic stenosis. Catheter Cardiovasc Interv. 2002; 56: 516–520.
Balmer C, Beghetti M, Fasnacht M, Friedli B, Arbenz U. Balloon aortic valvoplasty in paediatric patients: progressive aortic regurgitation is common. Heart. 2004; 90: 77–81.
Hawkins JA, Minich LL, Shaddy RE, Tani LY, Orsmond GS, Sturtevant JE, McGough EC. Aortic valve repair and replacement after balloon aortic valvuloplasty in children. Ann Thorac Surg. 1996; 61: 1355–1358.
Magee AG, Nykanen D, McCrindle BW, Wax D, Freedom RM, Benson LN. Balloon dilation of severe aortic stenosis in the neonate: comparison of anterograde and retrograde catheter approaches. J Am Coll Cardiol. 1997; 30: 1061–1066.(Doff B. McElhinney, MD; J)