ADVERTISEMENT
Late Outcomes after Pulmonary Valve Balloon Dilatation in Neonates, Infants and Children
June 2005
Since its initial description in 1982,1 percutaneous balloon dilation (PBD) has supplanted surgical valvotomy as the primary treatment modality for valvar pulmonary stenosis (PS) across all age groups. Short- and intermediate-term results are excellent,2–6 with reported 10-year freedom from re-intervention rates up to 85%. However, late outcomes with respect to pulmonary valve competence, right ventricular (RV) function and growth are not defined. In this study, we detail the determinants of late outcomes of PBD for valvar PS during childhood. Right heart growth patterns in relation to baseline characteristics and late sequelae were also assessed.
Methods
Study population. Children who had PBD for valvar PS between January 1984 and December 1992 were identified from the Hospital for Sick Children cardiology database. Children with associated small septal defects were included, whereas those with more complex anomalies were excluded. Hospital records were reviewed for demographics, baseline clinical features, procedural details and clinical status during follow-up. Right heart hemodynamics, before and after the procedure, pulmonary valve (PV) annular size (measured in the lateral projection from a right ventriculogram), and the presence of associated infundibular stenosis and/or valvar dysplasia were noted. Valve dysplasia was defined as poorly mobile leaflets, with abnormal and irregular valve thickening noted on the echocardiogram at catheterization7–9 or surgery. The association of a small PV annulus and absence of post-stenotic dilatation of the main pulmonary artery (MPA), together with valve dysplasia, were also noted. PBD was performed with a balloon 1.3 times the annular diameter.10–12 For larger annuli, a double balloon technique was used, the combined balloon diameters determined by a previously described formula.10 A single experienced observer (GV) reviewed all echocardiograms. Three studies were reviewed for each child (when available): before the procedure, at the first follow-up after the procedure, and at the latest follow-up. The following measurements were obtained from all studies: PV annular size (parasternal short axis view), tricuspid valve annular size at the level of the hinge points (4-chamber view), and MPA (at the midpoint between the PV and pulmonary arterial bifurcation). Z-score values (based on body surface area) were determined for individual measurements compared to normative data obtained from our laboratory. RV systolic dysfunction was visually graded as mild, moderate, or severe. Similarly, tricuspid regurgitation and pulmonary regurgitation (PR) were graded as trivial, mild, moderate, or severe, as described previously.13 For PR, the presence of reverse flow from the branch pulmonary arteries, in addition to a non-restrictive regurgitant Doppler signal across the RV outflow tract, was considered severe. Children without recent follow-up were contacted by telephone to verify live status, functional class, the need for re-intervention, and the presence of arrhythmias.
Statistics. Mean, standard deviation, median, and ranges were determined for continuous variables where appropriate; frequencies were determined for nominal and ordinal variables. Histograms were used to assess the distributions of continuous variables and to determine whether the parametric Student’s t-test or the non-parametric Wilcoxon rank-sum test was more appropriate for comparison between groups. A value of p = 0.8), the most significant factor was used in the multivariate regression model. The Kaplan-Meier method was used to describe freedom from re-intervention. Initial PBD was considered the index event. For children undergoing re-intervention, time to repeat balloon dilatation or surgical valvotomy was considered the duration of intervention-free follow-up. Patients having surgical re-intervention were censored at the time of operation, and were not included in late hemodynamic outcomes. The Cox regression model was used to determine factors associated with freedom from re-intervention.
Results
One-hundred and fifty patients were identified. Baseline demographic and clinical characteristics are outlined in Table 1. Of these, 26 were neonates, 8 having critical valvar stenosis, i.e., dependant on a prostaglandin E1 (PGE1) infusion to maintain pulmonary blood flow through a patent arterial duct. Three additional neonates were not on PGE1s, but were noted to have large ducts shunting left to right at the time of the procedure. Associated genetic anomalies included Noonan syndrome in 16 (11%), neurofibromatosis in 2, and 1 each with achondroplasia, William’s syndrome, and congenital rubella. Peri-procedural anatomic and physiologic characteristics are presented in Table 2. From the echocardiograms performed before the procedure, no child had more than mild TV regurgitation (18%). Similarly, no child had severe PR, 1 had moderate PR, and 15% had mild PR. Following the procedure, the mean PV gradient was reduced from 62.0 ± 29.2 mmHg to 22.8 ± 19.6 mmHg (p Re-interventions. Fifteen children, 5 with Noonan syndrome, had subsequent RV outflow tract surgery at a median age of 6 months (range; 2 days to 4.6 years). Of these, 11 children had a dysplastic PV with concomitant supravalvar stenosis in 6, and subvalvar stenosis in 4 children. Surgical procedures included pulmonary valvotomy in 14, RV outflow patch augmentation in 9 (including subvalvar muscle bundle resection in 2), and in 1 child who had RV hypoplasia, a bidirectional cavopulmonary shunt (Figure 1).
Of the 118 children not requiring early surgery, 11 had a second PBD at 3.7 ± 4.7 years (range = 6 days to 13.9 years) after the initial PBD, and 2 children had a third PBD for persistent PV obstruction. Two of these children required surgery; RV muscular bundles developed in 1 child, and infundibular stenosis in the other (Figure 1). There were no reported arrhythmias in any child at last follow-up, and only 2 were in NYHA class II (despite normal RV function and no significant PV gradient).
Two additional children underwent surgical repair for severe TV regurgitation at 11 and 12 years of age following PBD. At surgery, both were noted to have a flail anterior TV leaflet possibly secondary to a tear at the time of PBD. Overall freedom from any re-intervention rates at 1, 5, 10, and 15 years were: 90%, 83%, 83%, and 77%, respectively (Figure 2).
Early outcome. The first follow-up was after a mean of 0.9 ± 1.1 years. The PV gradient, as measured by echocardiography, was a mean of 26.2 ± 19.2 (n = 102). Moderate PR was seen in 22%, and severe PR in 2% (Table 3). Moderate TR was observed in 2% of children, and none had severe TR (n = 84). Table 4 delineates the growth of different right-sided structures at that time.
Late outcomes. The duration of follow-up was a mean of 11.9 ± 3.1 years (range = 3.7 to 19.3 years). While 16 children were lost to follow-up, their baseline characteristics were not different from the remainder of the study group. Pulmonary regurgitation was progressive during follow-up: at initial follow-up, 25 of 102 children (24%; with available echocardiograms) had moderate or severe regurgitation, whereas 54 of 95 children (57%) had moderate or severe PR at last follow-up (Table 3). Moderate TR was noted in 4 of 81 children (5%), and severe TR in only 1 (1%) child, not significantly different from the first follow-up. Two additional children had severe TR that was surgically repaired prior to the last follow-up. None of the univariate analysis done against different variables, for moderate-to-severe TR proved to be of significance.
From univariate analysis, children with moderate or severe PR at last follow-up had a smaller BSA at the time of intervention [0.52 vs. 0.82 m2 (p Neonatal group. Twenty-six children, with a mean weight of 3.7 ± 0.7 kg (range = 2.4 to 5.2 kg), were 2–6 Pulmonary regurgitation increases in severity during follow-up, and in the first decade, appears well-tolerated.
Early studies have shown that the mechanism of improved valve function after balloon dilatation is most often complete commissural, partial commissural, or peri-commissural tear.14 Thus, the procedure will be most successful when the obstruction is due to leaflet fusion, where the leaflets can be torn apart by the dilatation. This mechanism of leaflet disruption is the foundation for the adjunctive outcome of PR.
Those children with a smaller body surface area and those younger at the time of dilatation were more likely to develop moderate or severe PR during longer follow-up. In this regard, Berman et al.11 raised concern regarding the relative balloon over-sizing generally recommended in neonates and infants, and demonstrated progressive PR in 6 of their 107 patients undergoing PBD, 1 requiring pulmonary valve replacement at 7 years of age. In our series, PR was progressive in over half of the children during follow-up, and unrelated to the balloon-to-annular ratio. Our practice, based on previously reported series,10,15,16 has been to use a balloon-to-annulus ratio of 1.3:1, and therefore, it was not surprising that we were unable to demonstrate any significant correlations with the degree of regurgitation. Possible contributory factors for these observations are the longer follow-up duration in this series (up to 19 years), with the possible exacerbation of the anatomic perturbations such as leaflet tears or avulsions with time. Perforation of the valve is unlikely, but an irregular tear in those most immature valves (e.g. unicommissural)9 may contribute to the increasing PR.
Noonan syndrome may be associated with valvar PS in up to ~25% of cases.17 In patients with neurofibromatosis, cardiovascular lesions occur in approximately 2% of patients,18 the most common lesion being PV stenosis. In our series, associated genetic malformations including Noonan syndrome, achondroplasia, and neurofibromatosis were seen in 14% of children. Those with genetic syndromes commonly had associated valvar dysplasia, and in the Noonan group, frequently required surgical re-intervention to the right ventricular outflow tract. This finding of a lower success rate with a dysplastic valve has been noted in previous studies. In the study by DiSessa et al.,19 failure in the 3 balloon dilatations performed upon children with Noonan syndrome had a severely dysplastic valve. In contrast, Rao et al.20 described good results in balloon dilatation of dysplastic valves. Poor gradient reductions, but not utter failures, have been described in other studies.7,21 These differences are probably due to the amount of leaflet fusion that is associated with the dysplastic valve which can contribute to a partial response to dilatation. Additionally, the presence of a dysplastic valve syndrome complex will impact outcomes when present; the results are by far poorer. The mechanism of obstruction in these children, the bulky, non-mobile leaflets without leaflet fusion, and the small annulus with frequent supravalvar attachments, present multiple levels of outflow tract obstruction. Here, the leaflets are merely pushed aside by the balloon and fall immediately back in place without improving the hemodynamic outflow. In our series, 50% of the Noonan group were noted to have a dysplastic valve, but only 3 had associated features of complex outflow obstruction. As such, while the success rate was lower than the non-dysplastic group, it was still reasonably good enough to argue for offering this procedure as a first line treatment.
After a PBD, right heart structures appear to enlarge with time. The TV annulus grows appropriately, and the PV demonstrates a degree of catch-up growth in relation to patient body surface area. The MPA remodeled in follow-up, presumably reflects improvement in the hemodynamic burden on the vessel wall,22 with a reduction or elimination of the jet effect from the stenotic orifice. Importantly, children with moderate PR did not have significantly enlarged RV chambers. That this may be so is supported by evidence presented by D’Udeken et al.,23 who recently demonstrated the importance of infundibular function in preserving RV integrity in the presence of pulmonary regurgitation. The RV contracts as a bellows in a sequential fashion from diaphragmatic surface to outlet. The infundibulum is usually the last to contract and also the last to relax, thereby allowing early diastolic muscular restraint of potential regurgitant flow. In contrast to children with isolated PV stenosis after PBD, many children with repaired Fallot’s tetralogy have a transannular patch which not only disrupts infundibular function, but also directly damages RV myocardium, resulting in chamber dysfunction. This may lead to poor tolerance of PR, with associated RV decompensation. As such, moderate degrees of PR appear well-tolerated in these children with isolated pulmonary stenosis after PBD. Only those with severe PR resulted in significant RV dilatation. These findings underscore D’Udekem’s observations on infundibular function. No children in this series have as yet required PV replacement.
The freedom from re-intervention rate at 10 years was 83% in this series, and decreased to 77% at 15 years. Similar frequencies have been documented in earlier reports with shorter follow-up periods.2–4,6 Interestingly, in the series published by Gupta et al., there was a much lower re-intervention rate,24 perhaps due to the older age of this group, correlating with our finding of higher re-intervention rate if the procedure is performed at a younger age.
Conclusions
Long-term results of PBD during childhood are excellent. The right heart structures grow appropriately in the vast majority of children. Although late PR occurs commonly, it appears well tolerated in the first decade after intervention. For those children with severe PR, longer follow-up is necessary to determine if they develop RV decompensation and require valve replacement. Life-long follow-up is essential in children who have PBD for valvar pulmonary stenosis.
1. Kan JS, White RI Jr, Mitchell SE, Gardner TJ. Percutaneous balloon valvuloplasty: A new method for treating congenital pulmonary-valve stenosis. N Engl J Med 1982;307:540–542.
2. McCrindle BW, Kan JS. Long-term results after balloon pulmonary valvuloplasty. Circulation 1991;83:1915–1922.
3. Rao PS, Galal O, Patnana M, et al. Results of three to 10 year follow-up of balloon dilatation of the pulmonary valve. Heart 1998;80:591–595.
4. Tabatabaei H, Boutin C, Nykanen DG, et al. Morphologic and hemodynamic consequences after percutaneous balloon valvotomy for neonatal pulmonary stenosis: Medium-term follow-up. J Am Coll Cardiol 1996;27:473–478.
5. Jarrar M, Betbout F, Farhat MB, et al. Long-term invasive and noninvasive results of percutaneous balloon pulmonary valvuloplasty in children, adolescents, and adults. Am Heart J 1999;138:950–954.
6. McCrindle BW. Independent predictors of long-term results after balloon pulmonary valvuloplasty. Valvuloplasty and Angioplasty of Congenital Anomalies (VACA) Registry Investigators. Circulation 1994;89:1751–1759.
7. Musewe NN, Robertson MA, Benson LN, et al. The dysplastic pulmonary valve: echocardiographic features and results of balloon dilatation. Br Heart J 1987;57:364–370.
8. Jeffery RF, Moller JH, Amplatz K. The dysplastic pulmonary valve: a new roentgenographic entity; with a discussion of the anatomy and radiology of other types of valvular pulmonary stenosis. Am J Roentgenol Radium Ther Nucl Med 1972;114:322–339.
9. Gikonyo BM, Lucas RV, Edwards JE. Anatomic features of congenital pulmonary valvar stenosis. Pediatr Cardiol 1987;8:109–116.
10. Rao PS. Further observations on the effect of balloon size on the short term and intermediate term results of balloon dilatation of the pulmonary valve. Br Heart J 1988;60:507–511.
11. Berman W Jr, Fripp RR, Raisher BD, Yabek SM. Significant pulmonary valve incompetence following oversize balloon pulmonary valveplasty in small infants: A long-term follow-up study. Catheter Cardiovasc Interv 1999;48:61–65.
12. Radtke W, Keane JF, Fellows KE, Lang P, Lock JE. Percutaneous balloon valvotomy of congenital pulmonary stenosis using oversized balloons. J Am Coll Cardiol 1986;8:909–915.
13. Veldtman GR, Razack V, Siu S, et al. Right ventricular form and function after percutaneous atrial septal defect device closure. J Am Coll Cardiol 2001;37:2108–2113.
14. Burrows PE, Benson LN, Moes CA, Freedom RM. Pulmonary artery tears following balloon valvotomy for pulmonary stenosis. Cardiovasc Intervent Radiol 1989;12:38–42.
15. Robertson M, Benson LN, Smallhorn JS, et al. The morphology of the right ventricular outflow tract after percutaneous pulmonary valvotomy: Long term follow-up. Br Heart J 1987;58:239–244.
16. Syamasundar RP. Late pulmonary insufficiency after balloon dilatation of the pulmonary valve. Catheter Cardiovasc Interv 2000;49:118–119.
17. Burch M, Sharland M, Shinebourne E, et al. Cardiologic abnormalities in Noonan syndrome: Phenotypic diagnosis and echocardiographic assessment of 118 patients. J Am Coll Cardiol 1993;22:1189–1192.
18. Lin AE, Birch PH, Korf BR, et al. Cardiovascular malformations and other cardiovascular abnormalities in neurofibromatosis 1. Am J Med Genet 2000;95:108–117.
19. DiSessa TG, Alpert BS, Chase NA, et al. Balloon valvuloplasty in children with dysplastic pulmonary valves. Am J Cardiol 1987;60:405–407.
20. Rao PS. Balloon dilatation in infants and children with dysplastic pulmonary valves: Short-term and intermediate-term results. Am Heart J 1988;116:1168–1173.
21. Marantz PM, Huhta JC, Mullins CE, et al. Results of balloon valvuloplasty in typical and dysplastic pulmonary valve stenosis: Doppler echocardiographic follow-up. J Am Coll Cardiol 1988;12:476–479.
22. Veldtman GR, Dearani JA, Warnes CA. Low pressure giant pulmonary artery aneurysms in the adult: Natural history and management strategies. Heart 2003;89:1067–1070.
23. d'Udekem dY, Pasquet A, Lebreux L, et al. Does right ventricular outflow tract damage play a role in the genesis of late right ventricular dilatation after tetralogy of Fallot repair? Ann Thorac Surg 2003;76:555–561.
24. Gupta D, Saxena A, Kothari SS, Juneja R. Factors influencing late course of residual valvular and infundibular gradients following pulmonary valve balloon dilatation. Int J Cardiol 2001;79:143–149.
