Fetal pulmonary valvuloplasty for pulmonary atresia with intact ventricular septum: a single-center clinical experience

To the Editor: Fetal pulmonary atresia with intact ventricular septum (PA/IVS) or critical pulmonary stenosis (CPS) is a progressive disease from the time of mid-gestation diagnosis to birth. Elevated right ventricular (RV) afterload results in RV hypertrophy and reduced compliance, as well as a lack of flow through the right side of the heart, which leads to progressive RV and tricuspid valve growth arrest, and secondary damage of the RV myocardium. Studies have shown that tricuspid valve size, tricuspid valve Z-score, tricuspid regurgitation (TR), right ventricle dependent coronary circulation (RVDCC), and subaortic stenosis are all related to the two-ventricle postnatal strategy.[1] Most fetuses with PA/IVS or CPS with hypoplastic right heart syndrome (HRHS) will survive until birth, but with high morbidity and mortality during postnatal life.[2] Significant RV dysplasia may require multiple complex surgical strategies or other interventions during the patient's lifetime.[3] With a better understanding of the progression of congenital heart disease (CHD) and innovations in interventional techniques, intrauterine interventions have become possible for some CHD progressions. The goal of fetal pulmonary valvuloplasty (FPV) is to alter the natural progress of right-sided heart disease, or to prevent hemodynamic compromise and stimulate and promote prenatal RV growth, thus avoiding significant RV hypoplasia at birth. Prenatal treatment may be a unique opportunity to promote ventricular growth better than postnatal treatment. Linz Children's Hospital reported 23 fetuses with PA or CPS, showed that FPV has led to immediate larger RV caused by afterload reduction and filling increase, thus improving the chance for a biventricular (BV) outcome even in fetuses with a predicted non-BV circulation.[4] However, no specific data about pre-procedure dimensions or post-procedure RV growth have been reported. Therefore, the purpose of this study is to assess the development of RV size and function in utero, as well as the postnatal outcomes of FPV. As far as we know, this study described the largest sample size of FPV reported in China. The retrospective study was approved by the Ethics Committee of Women and Children's Hospital, Qingdao University (No. FELL-YX-2018-004). Written informed consent was obtained from all participants. All fetuses had an echocardiographic examination 1 to 3 days before and after FPV by the same experienced echocardiographer. Two-dimensional data were recorded to analyze retrospectively for ventricular and valvular dimensions and ratios, as well as tricuspid valve inflow time/cardiac cycle (TVI/CC) length, and TR velocity. Color Doppler flow image was used to describe the blood flow of ductus arteriosus and the level of TR. Criteria for FPV include either membranous atresia (rather than muscular) or critical stenosis of the pulmonary valve with a recognizable RV outflow tract (RVOT). The candidate's ventricular and valvular ratios and TVI/CC refer to the prediction standard of single ventricle circulation by Gómez. This scoring system is based on four parameters, such as, tricuspid/mitral annulus ratio (TV/MV) ≤ 0.83, right ventricle/left ventricle long-axis ratio (RV/LV) ≤ 0.64, and tricuspid valve inflow duration/cardiac cycle ratio (TVI/CC) (PV/AV) ≤ 0.75, tricuspid valve inflow duration/cardiac cycle ratio ≤ 0.365, with 100% sensitivity and 98% specificity predicting a non-BV outcome if three of the four criteria are fulfilled. After the diagnosis, we continued to observe the ventricular morphological developmental delay (2–3 weeks), complete retrograde flow in the ductus arteriosus, and severe TR, which is also our reference FPV indications. Exclusion criteria include muscular atresia of the RVOT, severe TR with low velocity (<2.5 m/s), severe fetal edema, or presence of large RV sinusoids. The most important aspect of FPV is the fetal position. The mothers underwent general anesthesia during FPV operation. Another important procedure is pre-procedural assembly of the balloon/wire unit before introduction of the trocar. With ultrasound confirmation of ideal trajectory to the pulmonary valve, the trocar (18 G or 19 G) is then advanced through the maternal abdomen, the uterus, amniotic cavity, the fetal chest, and into the RV outflow tract. The trocar should cross the pulmonary valve. The balloon is placed across the pulmonary valve and dilated with at least 80% of balloon to valve ratio. Technical success is defined as visualization of the balloon inflation across the valve and improved anterograde blood flow through the PV, with subsequent improvement in anterograde flow on color Doppler imaging with or without pulmonary regurgitation. Complications include pericardial effusion, tamponade, and bradycardia, which require immediate intervention with pericardiocentesis or intracardiac administration of epinephrine or atropine, respectively. Up to now, no major complications have been documented from the mother. Echocardiographic examination were repeated every 2 weeks after FPV. All live-born infants treated with FPV need intervention after birth. Children's clinical data were reviewed for postnatal procedures and outcomes. A BV outcome was defined as a circulation, where the RV was the only source of pulmonary blood flow in the absence of any signs of right heart failure. After 3 to 6 months of discontinuation of aspirin, ductal stent tended to close (<1 mm), and pulmonary circulation was stable without hypoxemia, which was also considered as BV circulation that could be achieved. An undetermined circulation as one where a modified Blalock–Taussig shunt or ductal stent was performed postnatally and followed up for <6 months or no discontinuation of anticoagulants, and the final circulation was identified as still unclear. FPV was carried out in 13 fetuses with PA/IVS (n = 10) or CPS (n = 3) from July 2018 to June 2019. The median gestational age at the time of intervention was 28 ± 4 weeks (24–32 weeks). Patients were from six different provinces in China. No fetal deaths were encountered and all fetuses achieved technical success after FPV. Pericardial effusions that needed treatment and persistent bradycardia occurred in 2/13 (15.4%) and 5/13 (38.5%) procedures, respectively. In the early stage of our development of FPV, we actively administered atropine to prevent bradycardia immediately after going through the trocar pathway. Two fetuses had severe persistent bradycardia after FPV, hence they were aborted. One patient was aborted because of RVDCC. Ten of 13 parameters were significantly improved, and the highest significance was observed for the RV inflow time. The measurement results before and after successful intervention are shown in Table 1. In the ten fetuses, the parameters measured by echocardiography were significantly improved, especially the RV inflow time. The measurement data before and after successful intervention are shown in Table 1. Progressive pulmonary valve stenosis and increased TR were observed in almost all fetuses during pregnancy. One case of re-atresia had been identified before birth by echocardiography. Table 1 - FPV postoperative indicators. Time point TV (cm) TV Z score TV/MV RV/LV PV/AV TVI/CC TR (m/s) Preoperative 0.84 ± 0.05 −0.08 ± 0.44 0.78 ± 0.17 0.67 ± 0.16 0.76 ± 0.15 0.36 ± 0.06 3.96 ± 0.49 2 weeks after FPV 0.87 ± 0.04∗ −0.66 ± 0.35∗ 0.94 ± 0.09∗ 0.84 ± 0.09∗ 0.96 ± 0.10∗ 0.39 ± 0.03∗ 3.34 ± 0.58∗ 6 weeks after FPV 0.92 ± 0.05† −1.59 ± 0.36† 0.97 ± 0.08† 0.86 ± 0.06 0.84 ± 0.18 0.40 ± 0.02 3.35 ± 0.67 Data are shown as mean ± standard deviation.∗Compared with the preoperative group, the difference was statistically significant (P < 0.05).†Compared with the 2 weeks after FPV group, the difference was statistically significant (P < 0.05). AV: Aortic valve; CC: Cardiac cycle length; FPV: Fetal pulmonary valvuloplasty; LV: Left ventricle; MV: Mitral valve; PV/AV: Pulmonary valve/aortic valve; PV: Pulmonary valve; RV: Right ventricle; TR: Tricuspid regurgitation; TV/MV: Tricuspid valve/mitral valve; TV: Tricuspid valve; TVI/CC: Tricuspid valve inflow time/cardiac cycle length; TVI: Tricuspid valve inflow time. There was significant difference in the TV, TV Z-score, TV/MV, RV/LV, PV/AV, TVI/CC, and TR between pre- and 2 weeks after FPV (t = −8.333, P < 0.001; t = 16.749, P < 0.001; t = −3.150, P = 0.012; t = −5.077, P = 0.001; t = −4.727, P = 0.001; t = −2.265, P = 0.04; t = 4.041, P = 0.003). Compared with 2 weeks after FPV, the level of TV, TV Z-score, and TV/MV were statistically significant in 6 weeks after FPV (t = 7.421, P < 0.001; t = −14.005, P < 0.001; t = 2.327, P = 0.045). The differences in the ratio of RV/LV, PV/AV, TVI/CC, and TR were not significant between 2 and 6 weeks after FPV (t = 0.474, P = 0.647; t = −1.809, P = 0.104; t = −1.253, P = 0.242; t = 0.030, P = 0.977). All the ten fetuses had no preterm birth, premature rupture of membranes, nor neonatal death. The median follow-up was 11.5 months (6–17 months). The median Z score of the tricuspid valve by postnatal echocardiogram was −1.57 (−1.71 to −0.83). About 6/10 of patients received percutaneous pulmonary valvuloplasty (PBPV) and patent ductus arteriosus (PDA) stenting at the same time after birth. The diameter of implanted PDA stent is 3.5 to 4.0 mm. Four of them were followed up for at least 10 months, and aspirin was stopped 3 months ago. The latest ultrasound result showed that PDA was <1 mm, which could be confirmed as BV circulation in the four cases. Two out of six patients were followed up for 6 months, and still with an undetermined circulation. Three out of ten patients were treated with PBPV alone, and one case did not receive any intervention. Ten neonates avoided surgical trauma in the early postnatal period. As a rare CHD, PA/IVS may progress to HRHS in late pregnancy. Fetuses with PA/IVS often manifest signs of severe hypoplasia and RV inadequacy and only receive single-ventricle palliation in the end.[5] For fetal cardiac intervention (FCI) to improve the natural progress of PA/IVS, the initial data is promising. Most patients undergoing FCI eventually achieve a BV outcome.[2,6] However, a significant number of PA/IVS patients may also progress to BV or 1.5 ventricular outcomes even without FCI. As a highly heterogeneous disease, PA or CPS with IVS are not among the factors associated with postnatal BV strategy.[7] Therefore, it is vital to develop the existing data carefully to predict who might need a complex fix, determining the best candidate for FCI, and getting the most benefit from FPV. The present study showed the natural course of PA/IVS in the uterus, as well as tricuspid valve annulus, mitral valve annulus, right ventricle length, and LV length diameters. For the entire cohort, the tricuspid valve Z score did not significantly change on average with advancing gestational age. Therefore, the candidate criteria for FPV in this study had been tricuspid valve Z score > −2.5 (median: −0.30, −0.36 to −0.27), characterized as mild or moderate RV hypoplasia. In this study, the intrauterine follow-up results after FPV showed that the fetal tricuspid valve development was slow and that no significant difference can be observed until 6 weeks after operation. In contrast, the tricuspid valve Z score relative to gestational age and fetal development demonstrated a trend of gradual aggravation after operation. However, combined with postnatal echocardiography, the tricuspid valve Z score was evaluated (−1.57 to −0.83). The Z scores of all children were in a mild range, despite some of them who were evaluated before operation having the single-ventricle outcome. TV, TV Z score, TV/MV, RV/LV, PV/AV, TVI/CC, and TR were significantly improved in the early postoperative period (within 2 weeks), indicating that the RV morphology and function improved more than before. This is consistent with the improved results of a previous study.[8] Six weeks after FPV, the raising of TV, TV Z score, and TV/MV continue to improve, which plays an important role in the long-term postnatal prognosis. The RV/LV and PV/AV did not improve, and even worsened. The larger RV diameter in the early stage after FPV was related to the decrease of afterload and increased filling. At the sixth week after operation, TVI/CC and TR had no significant improvement, and one child even suffered from atresia. Just as PBPV is accepted in postnatal pulmonary valve stenosis, the mechanism is still unclear. With consideration to the inner diameter of the puncture needle, it is difficult to achieve the complete opening of the atresia pulmonary valve by FPV. Therefore, a second FPV treatment still needs to be discussed, but it is difficult due to the decrease of amniotic fluid and the limitation of fetal position in the third trimester of pregnancy. Currently, we use a prediction score for a non-BV outcome as well as echocardiographic indications to screen candidate fetuses for FPV. Considering the current data in our center, FPV can significantly improve the structure and promote functional development of the RV, especially the development of tricuspid valve. Most of the live births will undergo BV circulation through intervention treatment. It should be pointed out that FPV fetuses with BV circulation are evaluated as a single ventricular outcome before operation. Funding This study was supported by grants from the National Natural Science Foundation of China (Nos. 81770316 and 81970249), Qingdao Science and Technology Plan (No. 20-3-4-47-nsh), and the Taishan Scholars Program of Shandong Province (2018). Conflicts of interest None.