Effect of Balloon Pulmonary Valvuloplasty on Growth of Pulmonary Annulus in Infants with Tetralogy of Fallot

Percutaneous balloon pulmonary valvuloplasty (PBPV) is an alternative intervention in infants with Tetralogy of Fallot (TOF). It can not only improve hypoxia but also promote pulmonary annulus (PA) growth. In this study, we evaluated the effect of PBPV on PA growth in infants with TOF. To eliminate the effect of the systemic to pulmonary shunt (SPS) that may promote PA growth, we divided TOF infants into 2 groups: group A, patients who underwent PBPV with or without other SPS, and group B, patients who attempted SPS but without PBPV. Sixty patients were included, 28 patients in group A and 32 patients in group B. Age at the time of intervention in group A (range, 0.4–5.4; median 1.4 months) was lower than that in group B (range, 2.3–7.7; median 4.8 months), p-value 0.02. The body weight in group A (range, 3–5.5; median 3.7 kg) was also lower than that in group B (range 4.1–6.4; median 5.9 kg), p-value 0.02. Echocardiographic data at the mean follow-up period of 37.2 months (3–88 months) in group A and 39.6 months (6–95 months) in group B demonstrated an increase in mean PA diameter from 5.0 ± 1.3 mm to 10.2 ± 2.9 mm, p-value < 0.001 in group A; and from 6.2 ± 2 mm to 9.5 ± 2.9 mm, p-value < 0.001 in group B. The median PA z-score increased from − 3.4SD (− 3.9 to − 2.6SD) to − 1.8SD (− 2.5 to − 0.8SD), with the p-value of 0.002 in group A; and increased from − 2.9SD (− 4.5 to − 1.3SD) to − 2.7SD (− 3.6 to − 1.4SD), with the p-value of 0.73 in group B. By using the PA z-score as the absolute value, there was a statistically significant increase in the PA z-score during follow-up in group A, but not in group B. Balloon pulmonary valvuloplasty in infants with TOF can facilitate the growth of the pulmonic annulus even after eliminating the effect of the systemic to pulmonary shunt.


Introduction
Tetralogy of Fallot (TOF) is the most common cyanotic heart defect, accounting for approximately 6.1% of all complex congenital cardiovascular diseases [1]. Definitive treatment of TOF is surgical correction, and the operative technique depends on the severity of the right ventricular outflow tract obstruction. Many studies suggest that 3-6 months of age and before one year of life is the optimal timing for elective corrective surgery in patients with TOF [2][3][4][5]. In neonates with symptomatic TOF, whether to perform complete repair or palliative procedure is still controversial [6,7]. However, Loomba et al. demonstrated a marked increase in the rate of transannular patch (TAP) in neonatal repair [8]. TAP often results in worse outcomes, for instance, increased right ventricular volume load [9], increased risk of reoperation [10][11][12], and increased rate of late sudden cardiac arrest [13].
One palliative procedure that has been performed in a patient with TOF is percutaneous balloon pulmonary valvuloplasty (PBPV). Previous studies have reported the safety [14][15][16] and efficacy of PBPV in terms of increased pulmonic annulus growth [14,16,17] and a decreased need for TAP at the time of complete repair at 6-12 months of age [18]. By contrast, in neonates with severe TOF even after PBPV, palliative systemic to pulmonary shunt (SPS) is still required in most patients [15]. Previous studies have also reported the effect of SPS on promoting pulmonary annulus (PA) growth [19,20].
We hypothesized that PBPV increased PA growth, which may result in a decreased risk of TAP at the time of total correction. The objective of this study was to evaluate the effect of balloon pulmonary valvuloplasty on PA growth in infants with Tetralogy of Fallot. Because SPS may also contribute to the growth of the pulmonary annulus, we compared PA growth in TOF infants between the 2 groups. Group A included patients who underwent PBPV with or without other SPS (either modified Blalock Taussig shunt (MBTS) or ductal stent (DS)). Group B included patients who attempted SPS (MBTS or DS) but did not undergo PBPV.

Patient
We reviewed the outpatient card document, catheterization report and operative record of patients with severe TOF, defined as those requiring prostaglandin E1 (PGE1) to maintain adequate oxygen saturation or having significant hypoxia based on the clinician's judgment at that time, who underwent a palliative procedure either PBPV or SPS to augment pulmonary blood flow before 1 year of age at King Chulalongkorn Memorial Hospital between 1 January 2007 and 31 December 2019. We excluded TOF patients with nonconfluent pulmonary artery branches, absent pulmonic valves, atrioventricular septal defects, and patients without imaging data. This study was approved by the Institutional Review Board (IRB), Faculty of Medicine, Chulalongkorn University (IRB No. 877/63).

Intervention
PBPV was performed under intravenous fentanyl and/or midazolam for sedation. Selective angiography was performed, and the pulmonary annulus was measured in lateral projection. The balloon size was selected according to the diameter of the pulmonary annulus, and the estimated balloon size to the diameter of the pulmonic valve ratio was approximately 1-1.2:1. The balloon dilation was performed by manual inflation until the release of the pulmonic valve restriction was seen.
The palliative systemic to pulmonary shunt in this study included MBTS and DS. MBTS was the main surgical technique. The mean MBTS size was 4.54 ± 0.73 mm (range, 3.5-6 mm). Ductal stenting was performed with bare or drug-eluting stents, and the mean stent diameter was 3.85 ± 0.49 mm (range, 2.5-4.5 mm).

Data collection
Demographic data from outpatient and inpatient records included age at intervention, sex, weight, height, prostaglandin E1 (PGE1) use, oxygen saturation and syndromic association. Complications and mortality related to the procedure were also recorded.
The transthoracic echocardiogram performed prior to intervention and at the last follow-up visit before total correction were reviewed in all patients. Data included the size of the pulmonary annulus, the pressure gradient across the right ventricular outflow tract, and the dimensions of the pulmonary arteries and the descending aorta for the calculation of the McGoon ratio and Nakata index. The measurement was performed by an unblinded pediatric cardiologist. The pulmonic annulus diameter and pulmonary artery size were measured from an inner-to-inner edge and in the parasternal short-axis view. The descending aorta diameter was measured in the subcostal long-axis view. The diameter of the pulmonary annulus was transformed into a z-score according to Detroit Data [21].

Statistical Analysis
The analysis was performed with STATA version 15. Descriptive analysis was presented as percentages, means and medians. The differences in mean or median between pre-and postintervention imaging data in the same group were assessed using a paired t-test or Wilcoxon signed rank test, respectively. The median difference (IQR) and the comparison between the mean or median difference between 2 groups were assessed using a two-sample independent t-test or Wilcoxon signed rank test, respectively. A p-value < 0.05 was considered statistically significant.

Baseline Characteristics
Seventy-five TOF patients were initially included. After exclusion of 15 patients, the total number of patients was 28 patients in group A and 32 patients in group B (Fig. 1).
The baseline demographic data are shown in Table 1. The median age at the time of intervention in group A (range 0.4-5.4; median 1.4 months) was lower than that in group B (range, 2.3-7.7; median 4.8 months), p-value 0.02. The body weight in group A (range 3-5.5; median 3.7 kg) was also lower than that in group B (range 4.1-6.4; median 5.9 kg), p-value 0.02. The level of right ventricular outflow tract obstruction in most patients was infundibular and valvular pulmonary stenosis, which did not show differences between the 2 groups. Concurrent SPS, whether MBTS or DS, was performed in 86% of PBPV patients; most were within the same admission. The rate of DS was higher in the PBPV group, approximately 50% in group A and 6.3% in group B, p-value < 0.001. The majority of DS (92%) were performed at the same time as PBPV. Other baseline characteristics, such as sex, PGE1 use, preprocedural oxygen saturation level, and associated syndrome, were not significantly different between the 2 groups.

Morbidity and Mortality
The complications and death at admission were not different between the 2 groups (Table 1). In group A, 1 patient had LPA stenosis due to an abnormal position of the ductal stent. In group B, 1 patient had chylothorax, 1 patient had pneumothorax, and 2 patients died after MBTS. One of these 2 patients had severe hypoxemia and hemodynamic instability during the MBTS operation requiring a switch to a central shunt; however, the patient had unstable conditions that led to sudden cardiac arrest after the operation. The other patient developed bradycardia and sudden  The PBPV group showed a higher re-intervention rate than the non-PBPV group (Table 1). Early re-intervention occurred in 6 patients in the PBPV group; 4 patients who had only PBPV needed SPS to improve oxygenation at 6-51 days after the initial procedure and 2 patients needed balloon dilatation of DS and PA branches. For the late reintervention in the PBPV group; 13 patients had balloon dilatation of DS with or without dilatation of stenotic PA branches, 5 patients had additional MBTS placement and 1 patient had repeated PBPV. For the non-PBPV group, 7 patients had balloon dilatation of previous MBTS or DS, 3 patients had additional MBTS placement and 1 patient with previous MBTS had additional DS.

Pre-and Postintervention Echocardiographic Parameters
From the preintervention echocardiographic parameters, the initial PA size was 5 ± 1.3 mm in group A and 6.2 ± 2 mm in group B. The median PA z-score was − 3.4 SD (− 3.9 to − 2.6 SD) and − 2.9 SD (− 4.5 to − 1.3 SD) in groups A and B, respectively. The mean follow-up time after the first intervention was 37.2 months (3-88 months) in group A and 39.6 months (6-95 months) in group B. After intervention, the pulmonary annulus in both groups showed a significant increase in size compared with preintervention. However, when calculated as the PA z-score standardized by the patient's weight and height, the median PA z-score showed significant growth in group A but not in group B, as shown in Table 2 and Fig. 2. Most of the patients who underwent PBPV achieved a PA z-score greater than -2 at the last follow-up visit. The size of the pulmonary artery, McGoon ratio, and Nakata index increased significantly in both groups (Table 2). Table 3 demonstrates a comparison of the mean difference in the PA dimension, pulmonary artery size, McGoon ratio and Nakata index before and postintervention between the two groups. Group A showed a significant increase in the pulmonary annulus diameter and z-score after intervention compared with group B. Figure 3 shows a comparison of the change in PA size between the two groups. There was no significant difference in pulmonary artery size, McGoon ratio, or Nakata index after intervention between the two groups.

Discussion
Balloon pulmonary valvuloplasty is a palliative procedure in TOF patients. Many studies have reported on the safety and favorable outcome of this procedure [14-16, 18, 22]. We found a complication rate of 3.5%, which was comparable with other studies [23]. Several previous studies have reported a favorable effect of PBPV performed in TOF patients from 3 days to 3 years old on PA growth, a significant increase in PA z-score after intervention [16,18,22] and a 30-40% reduction in the transannular patch at the time of total repair [18]. Furthermore, a recent study reported greater PA growth and a lower need for TAP in patients who underwent PBPV than in those who underwent SPS and infundibulectomy [24]. Cholkraisuwat et al. [14] also reported PA growth from a z-score of − 2.56 to − 1.87 Table 2 Comparison of echocardiographic data between pre-and postintervention in groups A and B Differences in mean or median between pre-and postintervention were assessed using a paired t-test or Wilcoxon signed rank test, respectively PA pulmonary annulus, RPA right pulmonary artery, LPA left pulmonary artery, DAO descending aorta * P-value was evaluated using McNemar's test (p-value < 0.05) in 51 TOF patients who underwent PBPV at a median age of 3 years and 5 months. The explanation is quite straightforward in that PBPV increases forward flow across the pulmonary valve, resulting in significant PA growth. Our study reported a similar result that the PA diameter and z-score significantly increased in TOF patients who underwent PBPV during the infancy period (mean PA diameter increased from 5 to 10.2 mm and median PA z-score increased from − 3.4 to − 1.8).
Nakashima et al. [19] and Chong et al. [20] proposed a hypothesis that systemic to pulmonary shunts can facilitate pulmonary annulus growth and may be a confounding factor that contributes to an increase in PA diameter and z-score rather than the effect of PBPV. This finding could be explained by increased LV volume loading after SPS, which may increase forward flow through the pulmonary valve, thus increasing PA growth. The other explanation may be from increased pulmonary pressure resulting in dilatation of the pulmonary annulus. Most PBPV patients in our study needed concurrent SPS, either MBTS or DS, mainly within the same admission to maintain adequate oxygen saturation. This indicated a significant degree of disease severity. A small pulmonic annulus at baseline could not grow fast enough immediately after PBPV; hence initial PBPV alone could not provide sufficient pulmonary blood flow so additional SPS was required to maintain optimal oxygen saturation in this group of patients. To eliminate the effect of SPS, we compared the pulmonic valve growth and z-score between patients who underwent PBPV and SPS and patients who underwent SPS without PBPV. We found that PA diameter had significant growth following somatic growth in the patients in both study groups. However, when  calculated as the PA z-score, patients who underwent PBPV had a significant increase in the PA z-score during follow-up. In contrast, patients who underwent SPS alone did not show a significant increase in the PA z-score after the intervention. This would be a significant finding that emphasized the favorable effect of PBPV in promoting pulmonary annulus growth, apart from SPS itself. Furthermore, attention should be paid to the PA z-score rather than the PA diameter in the follow-up of this group of patients. Regarding the type of SPS, DS was performed more in the PBPV group. One of the explanations would be that it could be performed within the same procedure as PBPV. According to one large systematic review and meta-analysis [25], there was no significant difference in pulmonary artery growth between the ductal stenting and surgical systemic to pulmonary shunt for ductal-dependent pulmonary circulation. Therefore, the different types of SPS may not contribute to a difference in pulmonic annulus growth in this study.
The early re-intervention rate was higher in the PBPV group mainly because the patients who had only PBPV initially presented with hypoxia and needed SPS to supply adequate pulmonary blood flow within 2 months. Late reintervention also occurred more in the PBPV group mainly related to DS re-intervention. According to the late repair in our center, DS may become stenotic over time requiring further re-inventions to improve stent patency. A similar number of patients (27%) in both groups needed additional SPS placement while waiting for the total repair.
The current strategy of surgical correction in TOF aims to decrease the transannular patch according to unfavorable long-term outcomes that have been reported in many studies [9][10][11][12][13]. The pulmonary annulus z-score is one of the contributing factors associated with the rate of TAP at the time of total repair. The increase in PA z-score will result in a decreased risk of TAP in this group of patients [26,27]. Although there are a variety of PA z-scores used as cut-off points for TAP placement, a z-score greater than − 2 seems to be reasonable and has been used in many centers [28]. In this study, the number of patients who had a PA z-score greater than − 2 increased from 15 to 60% at the mean follow-up time of 37.2 months in the PBPV group but did not increase in the non-PBPV group, as shown in Table 2. This implies that PBPV may reduce the need for TAP at the time of total correction. However, the rate of TAP placement will be assessed after the patients in this study have undergone surgical total repair.

Study Limitation
Due to the study's retrospective design, several missing data may affect data analysis and outcome. The median age of the patients between the two groups was also different but was still in the early infancy period. Whether the age group difference contributed to a different outcome may be beyond our analysis in this study. Furthermore, the age for the total repair of TOF was quite late in our center, with the median age of repair being 4.9 years (IQR 3.4-8.2 years), according to the limited resource regarding available operative schedules and the number of surgeons. Only a few patients who had interventions in this study underwent total repair, so we only preliminarily reported a promising growth in the pulmonic annulus. However, we could not conclude whether this would affect the rate of a transannular patch at the time of total repair. We will continue to follow this group of patients until all undergo total repair.

Conclusion
Balloon pulmonary valvuloplasty in infants with TOF can facilitate the growth of the pulmonic annulus even after elimination of the systemic to pulmonary shunt effect. The significant increase in pulmonic valve annulus diameter after PBPV may reduce the need for transannular patches at the time of total correction. Author Contributions All authors contributed to the manuscript as follow:Substantial contributions to the conception or design of the work (KW, SL, PL), acquisition of data (KW) or analysis and interpretation of data (KW, SL, PL). Drafting the work (KW) or revising it critically for important intellectual content (SL, PL). Final approval of the version to be published (KW, SL, PL, JN, VB). Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved (KW, SL, PL, JN, VB).