Abstract
Background: We evaluated the surgical outcomes of modified right ventricle (RV) overhaul (mRVOh), implemented as part of comprehensive management for pulmonary atresia with intact ventricular septum (PA-IVS).
Methods and Results: Twenty-five mRVOh procedures were performed in 23 patients with PA-IVS without RV-dependent coronary circulation. The procedure involved RV sinus myectomy, infundibular muscle resection, and tricuspid valve (TV) and pulmonary valve (PV) repair. In addition, in neonates and young infants, Blalock-Taussig shunt or patent ductus arteriosus banding was performed simultaneously. TV and PV annulus sizes were measured using echocardiography; RV function and volume were assessed using magnetic resonance imaging (MRI) in 18 patients. The median age and body weight at the time of mRVOh were 7.0 months and 7.1 kg, respectively. Biventricular repair was performed in 19 patients, and 6 required reoperations, including 2 with redo mRVOh. After mRVOh, the mean TV and PV annulus z-scores showed a significant increase towards the normal range, from −1.91 to −1.40 (P=0.031), and from −2.23 to −1.11 (P=0.014), respectively. Serial postoperative MRI showed significant increases in RV end-diastolic and end-systolic volume indices, stroke volume index, and cardiac index (P<0.001 for all), with preserved RV function.
Conclusions: Both RV size and TV annulus showed proportionate growth after mRVOh. mRVOh may be a viable option for facilitating sustainable RV and TV growth in selected patients with PA-IVS.
Central Figure
The initial goal of treating pulmonary atresia with intact ventricular septum (PA-IVS) is to decompress the right ventricle (RV) and establish adequate antegrade pulmonary blood flow through the RV to ultimately promote RV growth for biventricular repair (BVR).1–7 Although single-ventricle palliation (SVP) is generally chosen in patients with RV-dependent coronary circulation (RVDCC) or severe hypoplasia of the RV and tricuspid valve (TV), the optimal choice between BVR and SVP for PA-IVS remains controversial, particularly in patients with borderline RV hypoplasia related to a smaller TV.2,4–13
Previous studies have suggested various treatment strategies for PA-IVS. However, attempts to promote RV growth, including traditional RV overhaul (RVOh), have consistently shown an increase in RV size, whereas the TV annulus size increased disproportionately relative to RV growth.2–4,6,10,13 Therefore, even successful BVRs for PA-IVS may not guarantee proportional growth of the TV annulus relative to adequate RV growth. In this context, we hypothesized that resecting hypertrophied RV muscles and endocardial fibrotic tissue at an early stage may facilitate RV and TV growth, thereby maximizing their growth potential through early RV optimization. Therefore, we have adopted a modified RVOh (mRVOh) procedure, involving RV sinus myectomy and pulmonary valvotomy under cardiopulmonary bypass (CPB) during the initial surgical palliation for RV decompression. This surgical one-stage primary repair strategy, aimed at optimizing RV and TV growth, yielded favorable outcomes, including significant increase in TV annulus size during postoperative follow-up.14,15 Building on our previous research, we conducted a series of further studies with extended follow-up, using magnetic resonance imaging (MRI) to quantitatively measure RV size and precisely assess RV growth trends.
The aim of this study was to comprehensively review and evaluate the clinical outcomes of mRVOh, including RV sinus myectomy (RVSM). We sought to establish whether mRVOh promotes effective growth in right-sided cardiac structures through quantitative assessment of RV size and function using MRI during follow-up and to assess long-term outcomes following mRVOh.
Methods
Ethics Statement
The Institutional Review Board of Seoul National University Hospital reviewed the study protocol and approved the present study as a minimal-risk retrospective study (Approval no. H-2402-100-1512; approval date, March 8, 2024) that did not require individual patient consent on the basis of the institutional guidelines for waiving consent.
Patients
Sixty patients diagnosed with PA-IVS underwent treatment at Seoul National University Children’s Hospital between January 2008 and August 2023. Balloon pulmonary valvuloplasty (BPV) was initially attempted in all patients at Seoul National University Children’s Hospital or in other institutions. Fifteen patients were successfully treated with BPV alone, whereas the remaining 45 subsequently required surgical treatment for definitive repair at Seoul National University Children’s Hospital. Of these 45 patients, 22 did not undergo the mRVOh procedure either due to the presence of RVDCC (n=8) or extreme RV size without RVDCC (n=14; specifically, 9 with nearly normal RV size [TV annulus z-score ≥0] and 5 with severe RV hypoplasia [TV annulus z-score <−4.0]). The remaining 23 patients without RVDCC, deemed candidates for BVR, underwent mRVOh.
The mRVOh selection criteria included borderline RV hypoplasia (TV annulus z-score between −4.0 and −2.0). In addition, patients with a TV annulus z-score ≥−2.0 were selected for RVOh if they had persistent right ventricular outflow tract (RVOT) obstruction despite BPV or exhibited an insufficient RV growth trend over a minimum 3- to 6-month follow-up period after BPV. RVOh was considered for these patients to optimize RV growth for successful BVR and to mitigate the risk of RV failure progression. Finally, this study included 25 consecutive mRVOh procedures performed in 23 patients (Figure 1).
Among the participants, 12 patients has been referred to Seoul National University Children’s Hospital from other hospitals after undergoing interventions after birth, 3 were referred with a one-and-a-half ventricle status, and none had undergone RVSM prior to mRVOh. Of the 11 patients initially treated at Seoul National University Children’s Hospital, 5 received initial palliation before mRVOh.
Patients with suspected RVDCC based on the initial echocardiographic evaluation underwent coronary angiography for confirmation. Computed tomography angiography was performed as an alternative in patients deemed unsuitable for invasive procedures. In patients in whom RVDCC could not be definitively excluded based on preoperative evaluation, intraoperative assessment was conducted before initiating CPB. The surgical strategy was finalized based on these findings. This approach ensured the exclusion of patients with RVDCC and a more accurate patient selection for mRVOh.
Operative Strategy and Surgical Procedures
To optimize RV growth in PA-IVS, we performed mRVOh in patients deemed suitable for BVR. Consistent with this approach, we previously outlined our operative strategies and techniques aimed at maximizing the potential for effective RV growth in these patients (Figure 2).14–16 Endocardial fibrotic tissue in the inlet, trabecular, and outlet portions of the RV may worsen RV compliance and diastolic function, thereby potentially impeding effective RV growth. Thus, RVSM and endocardial fibrotic tissue resection were performed simultaneously in the presence of these tissues. RVSM was first performed through the TV, followed by pulmonary valvotomy, and completed with resection of residual hypertrophied muscles through the pulmonary valve (PV). This comprehensive approach avoids right ventriculotomy while preserving long-term RV function. Valvuloplasty for the PV or TV was performed as needed, and hypertrophied RV infundibular muscle resection through the PV was routinely performed after pulmonary valvotomy.
In neonates, if BPV fails as the primary intervention for RV decompression, surgery is required. In addition, surgery was considered if BPV did not achieve adequate pulmonary blood flow because of severely hypertrophied RV infundibular muscles or ineffective BPV. For postintervention patients referred from other hospitals, the treatment strategy was determined based on their status. For patients with a history of previous BPV or surgery, such as the Blalock-Taussig shunt (BTS) or BTS followed by bidirectional cavopulmonary shunt (BCPS), mRVOh was performed based on the patient’s current status.
For mRVOh in neonates and young infants, we prefer patent ductus arteriosus (PDA) banding to BTS to ensure adequate pulmonary blood flow. PDA banding was performed using a polytetrafluoroethylene vascular graft strip, wrapped around the ductus with an applied width of approximately 3–4 mm. Under echocardiographic guidance, the tightness of the band was adjusted by clipping the strip to achieve an internal diameter approximately equal to the patient’s body weight in millimeters. Fixation sutures were placed on the adventitia of the pulmonary artery to prevent migration of the polytetrafluoroethylene strip. Following PDA banding, prostaglandin E1
was administered for approximately 2 weeks to maintain ductus patency. The next step, either discontinuing prostaglandin E1
to facilitate spontaneous closure or placing a PDA stent, was determined by the adequacy of antegrade pulmonary blood flow through the RVOT.
Atrial-level shunt management and one-and-a-half ventricle repair decisions were based on multidisciplinary discussions and intraoperative echocardiographic findings. The feasibility of biventricular conversion was assessed using cardiac MRI during follow-up. Atrial-level shunts were managed by either partially closing the atrial septal defect (ASD) without a patch or with a fenestrated patch, or by preserving the patent foramen ovale, based on the patient’s condition. During follow-up, spontaneous or device closure for residual ASD was contingent on patient status.
Finally, to ensure healthy and sustainable RV growth while preserving its long-term function, we aimed to minimize the risk of RV enlargement, hypertrophy, or both, which may occur following issues involving the TV and PV. Therefore, we actively addressed TV- and PV-related issues to support optimal and sustained RV growth.
Echocardiographic Evaluation
All preoperative and postoperative echocardiographic data were obtained and subsequently reviewed by 2 cardiologists (S.M.B., H.W.K.), who validated the final findings. Annulus size and the valvular function of the TV and PV were assessed using transthoracic echocardiography, with z-scores calculated based on body surface area.17 Perioperative and follow-up valvular regurgitation were classified as follows: 0, none to trivial; 1, mild; 2, mild to moderate; 3, moderate; and 4, severe.18 RV morphology was classified as tripartite, bipartite, or unipartite based on the development of the 3 RV components.19
MRI Assessment
During follow-up, MRI scans were performed twice on 18 of the 23 patients using a 1.5-T unit (Avanto; Siemens, Erlangen, Germany). Of these patients, 2 who underwent mRVOh twice did not undergo MRI after the initial procedure. The median interval between scans was 45.4 months (interquartile range [IQR] 17.2–56.1 months). Our cardiac MRI protocol included thoracic survey images captured from axial, coronal, and sagittal angles, cine sequences using steady-state free precession in 2- and 4-chamber views, RVOT view, and short-axis stacks spanning from the base to the apex. The cine sequence parameters were as follows: slice thickness, 6 mm; field of view, 300 mm×300 mm; matrix, 240×216; and temporal resolution, 45.63 ms. Analysis was conducted using the Argus post-processing tool (Siemens Medical Solutions). RV and left ventricle (LV) endocardial contours were obtained from the short-axis view in a semi-automated manner, excluding papillary muscles and trabeculae from the cavity. The RV and LV systolic and diastolic volumes were determined using Simpson’s rule.
Follow-up and Evaluation of Clinical Outcomes
Postoperative clinical and echocardiographic follow-up was conducted at 3- to 6-month intervals for the first 2 years and annually thereafter in all patients. Data on survival status and the presence of cardiovascular events were collected from electronic medical records. Information on reoperation and the occurrence of major adverse events was collected. The median clinical and echocardiographic follow-up durations were 6.12 years (IQR 4.63–8.53 years) and 5.25 years (2.23–8.28 years), respectively. Operative mortality was defined as death within 30 days of surgery or during the same hospitalization. Late mortality was defined as death after discharge. Significant RV dysfunction was defined as an RV ejection fraction <40% on cardiac MRI.
Statistical Analysis
Continuous variables are expressed as the mean±SD or median with IQR, and categorical variables are presented as frequencies. Changes in continuous outcomes were assessed using multiple linear regression, adjusting for the period from the preoperative period to the last follow-up or from the first to the second MRI. Changes in the grade of tricuspid regurgitation (TR) and pulmonary regurgitation (PR) were analyzed with an ordinal logistic regression model using the generalized estimating equation for repeated ordered responses (preoperative and follow-up TR and PR grade) to adjust for the follow-up period. Reoperation rates were estimated using the Kaplan–Meier method. Cox proportional hazards regression, logistic regression, and linear regression were used to analyze factors associated with reoperation, BVR, and continuous outcomes (TV and PV annulus z-scores, as well as cardiac volume and function), respectively.
Statistical analyses were performed using IBM SPSS version 25.0 (IBM Inc., Armonk, NY, USA) and SAS for Windows version 9.4 (SAS Institute, Cary, NC, USA). All P values are 2-tailed, with significance set at P<0.05.
Results
Baseline Characteristics and Perioperative Data
The baseline characteristics and perioperative data are summarized in Table 1. The median age and body weight at the time of mRVOh were 7.0 months (IQR 0.7–20.5 months) and 7.1 kg (IQR 3.6–12.5 kg), respectively. Seven patients were neonates (age ≤30 days), and 3 were young infants (age <2 months). Fourteen patients had a bipartite RV, and 2 had a major RV-coronary artery fistula.
Table 1.
Patient Baseline Characteristics and Perioperative Data (n=23 Patients)
| Age at the time of surgery (months) |
7.0 [0.70, 20.53] |
| Body weight at the time of surgery (kg) |
7.1 [3.6, 12.5] |
| Body surface area (m2) |
0.35 [0.23, 0.54] |
| Sex |
| Male |
16 (69.6) |
| Female |
7 (30.4) |
| Preterm (gestational age <37 weeks) |
2 (8.7) |
| Low birth weight (<2.5 kg) |
3 (13.0) |
| Genetic syndrome |
1 (4.4) |
| Neonate (age ≤30 days) |
7 (30.4) |
| Young infant (age <2 months) |
3 (13.0) |
| RV morphology |
| Bipartite |
14 (60.9) |
| Tripartite |
9 (39.1) |
| Major RV-to-coronary artery fistula |
2 (8.7) |
| Preoperative interventions at SNUCH |
11 (47.8) |
| None |
6 (26.1) |
| Transcatheter±surgical |
5 (21.7) |
| Preoperative interventions at other institutions |
12 (52.2) |
| Only transcatheter |
4 (17.4) |
| Only surgical |
1 (4.4) |
| Both |
7 (30.4) |
| Preoperative one-and-a-half ventricle status |
3 (13.0) |
| Preoperative TV annulus z-score |
−1.94 [−3.09, −1.19] |
| Preoperative tricuspid regurgitation grade |
| None-to-trivial |
7 (30.4) |
| Mild |
6 (26.1) |
| Mild-to-moderate |
1 (4.4) |
| Moderate |
3 (13.0) |
| Severe |
6 (26.1) |
| Preoperative PV annulus z-score |
−1.92 [−3.12, −1.29] |
| Preoperative pulmonary regurgitation grade |
| None-to-trivial |
12 (52.2) |
| Mild |
5 (21.7) |
| Mild-to-moderate |
1 (4.4) |
| Moderate |
3 (13.0) |
| Severe |
2 (8.7) |
| Preoperative pulmonary stenosis |
| Not significant (<36 mmHg) |
4 (17.4) |
| Significant (≥36 mmHg) |
19 (82.6) |
| Concomitant procedures |
| PV repair/replacement |
22 (95.7) |
| Pulmonary valvotomy |
21 (91.3) |
| PV replacement |
1 (4.4) |
| Repair of BPV-related PV leaflet injury |
9 (39.1) |
| TV repair |
10 (43.5) |
| Papillary muscle splitting |
5 (21.7) |
| Secondary chordae resection |
5 (21.7) |
| Commissurotomy |
3 (13.0) |
| Leaflet augmentation |
2 (8.7) |
| Artificial chordae implantation |
2 (8.7) |
| Annuloplasty |
1 (4.4) |
| Atrial septal procedure |
17 (73.9) |
| Fenestrated patch closure |
13 (56.5) |
| Patent foramen ovale preservation |
4 (17.4) |
| Patent ductus arteriosus banding |
6 (26.1) |
| Cardiopulmonary bypass time (min) |
141.0 [106.0, 164.0] |
| Aortic cross-clamp time (min) |
81.0 [60.0, 107.0] |
Values are presented as the median [interquartile range] for continuous data or as n (%) for categorical data. BPV, balloon pulmonary valvuloplasty; PV, pulmonary valve; RV, right ventricle; SNUCH, Seoul National University Children’s Hospital; TV, tricuspid valve.
Of the 12 patients referred to Seoul National University Children’s Hospital, 4 underwent BPV only and 8 underwent surgical interventions (7 underwent BTS following BPV, and 1 underwent only BTS). Of the 11 patients initially treated at Seoul National University Children’s Hospital, 4 underwent BPV only and 1 underwent BTS with RVOT patch widening following BPV prior to mRVOh. Overall, 17 patients underwent transcatheter (n=16) and/or surgical (n=9) intervention for PA-IVS before mRVOh, including 1 who underwent RV myectomy during a prior surgical intervention.
Clinical Outcomes
There were no operative mortalities and postoperative morbidity included respiratory complications following prolonged mechanical ventilation (n=2) and mediastinitis (n=2). No cases of late mortality occurred during follow-up; however, 6 patients required reoperation for TV and PV repairs due to significant regurgitation or stenosis. In addition, 2 patients underwent mRVOh twice, as deemed necessary during follow-up, primarily to promote further RV growth, with additional procedures performed to address associated lesions as needed. One of these patients, who underwent mRVOh at 20 days of age, required reoperation at 28.2 months of age for significant TR and pulmonary stenosis (PS) with a peak pressure gradient of 44.6 mmHg. This procedure included ASD closure and PDA division in addition to redo mRVOh. The other patient underwent reoperation at 28.2 months of age due to right heart failure and significant PR after the initial mRVOh at 70 days of age. We performed BCPS, ASD patch closure with fenestration, and PDA division, alongside mRVOh. The reoperation-free rates at 1, 3, and 5 years were 95.7%, 81.5%, and 75.7%, respectively.
No ventricular arrhythmias were observed during follow-up. Although postoperative right bundle branch block was identified in 14 patients, all remained asymptomatic and required no further intervention. The final cardiac MRI showed no evidence of significant RV dysfunction. Furthermore, no echocardiographic findings suggestive of RV dysfunction were observed on the last echocardiogram in patients who did not undergo cardiac MRI.
At the last follow-up, 19 (82.6%) of the 23 patients had achieved BVR, whereas the remaining 4 remained in a one-and-a-half ventricle repair state. Of these 4 patients, 2 underwent concomitant BCPS: 1 during the initial mRVOh and the other during the redo mRVOh. Of the 3 patients who had undergone one-and-a-half ventricle repairs in other institutions before undergoing the mRVOh procedure, one eventually achieved successful biventricular conversion. The atrial-level shunt left during the procedure closed spontaneously in 9 of the 17 patients, whereas atrial-level shunt flow was observed in the remaining patients.
Changes in TV and PV on Echocardiographic Evaluation
During follow-up, the mean TV annulus z-score increased significantly from −1.91±1.53 preoperatively to −1.40±1.10 at the last follow-up (P=0.031). The mean PV annulus z-score also showed a significant increase from −2.23±1.73 preoperatively to −1.11±0.92 at the last follow-up (P=0.014). During follow-up, significant (≥moderate) TR developed in 4 (17.4%) patients and significant (mean pressure gradient >5 mmHg) tricuspid stenosis was observed in only 1 patient. Significant (≥moderate) PR occurred in 5 (21.7%) patients, and only 1 (4.4%) patient had significant (peak velocity ≥3 m/s) PS, requiring BPV for moderate PS after surgery. The echocardiographic data are summarized in Table 2 and Figure 3.
Table 2.
Echocardiographic Changes in Size and Function of the TV and PV
| |
Preoperative
(n=23) |
At the last follow-up
(n=23) |
Estimate |
SE |
P value |
| TV annulus z-score |
|
|
0.474 |
0.205 |
0.031 |
| Mean±SD |
−1.91±1.53 |
−1.40±1.10 |
|
|
|
| Median [IQR] |
−1.94 [−3.09, −1.19] |
−1.49 [−2.08, −0.57] |
|
|
|
| TV annulus (% of predicted normal*) |
|
|
9.431 |
2.308 |
<0.001 |
| Mean±SD |
79.93±16.32 |
89.65±9.32 |
|
|
|
| Median [IQR] |
78.73 [67.77, 91.56] |
89.97 [86.94, 96.36] |
|
|
|
| PV annulus z-score |
|
|
1.066 |
0.398 |
0.014 |
| Mean±SD |
−2.23±1.73 |
−1.11±0.92 |
|
|
|
| Median [IQR] |
−1.92 [−3.12, −1.29] |
−1.32 [−1.84, −0.38] |
|
|
|
| PV annulus (% of predicted normal*) |
|
|
20.319 |
4.528 |
<0.001 |
| Mean±SD |
68.71±20.11 |
89.53±9.43 |
|
|
|
| Median [IQR] |
69.40 [52.88, 85.17] |
88.65 [83.31, 95.56] |
|
|
|
| RVOT pressure gradient (mmHg) |
|
|
−31.849 |
9.757 |
0.008 |
| Mean±SD |
47.28±31.74 |
16.46±9.41 |
|
|
|
| Median [IQR] |
46.0 [6.8, 116.6] |
16.0 [5.2, 44.55] |
|
|
|
| |
|
|
OR |
95% CI |
P value |
| Tricuspid regurgitation |
|
|
0.682 |
0.370–1.257 |
0.220 |
| None-to-trivial |
7 (30.4) |
4 (17.4) |
| Mild |
6 (26.1) |
13 (56.5) |
| Mild-to-moderate |
1 (4.4) |
2 (8.7) |
| Moderate |
3 (13.0) |
3 (13.0) |
| Severe |
6 (26.1) |
1 (4.4) |
| Pulmonary regurgitation |
|
|
3.238 |
1.291–8.118 |
0.012 |
| None-to-trivial |
12 (52.2) |
3 (13.0) |
| Mild |
5 (21.7) |
9 (39.1) |
| Mild-to-moderate |
1 (4.4) |
6 (26.1) |
| Moderate |
3 (13.0) |
4 (17.4) |
| Severe |
2 (8.7) |
1 (4.4) |
*The measured valve annulus diameter expressed as a percentage of the expected normal value for body surface area (corresponding to z-score=0). CI, confidence interval; IQR, interquartile range; OR, odds ratio; PV, pulmonary valve; RVOT, right ventricular outflow tract; SE, standard error; TV, tricuspid valve.
In 2 patients undergoing redo mRVOh, the TV annulus z-scores increased from 0.29 and −1.82 at initial surgery to 0.39 and −1.68, respectively, at redo-surgery. At the last follow-up, the z-scores were −0.55 and 0.17, respectively.
Comparative Quantitative Assessment of RV Size and Function on Serial Cardiac MRI
The median age at the first and second cardiac MRIs was 2.31 years (IQR 1.49–5.07 years) and 6.71 years (IQR 5.16–9.70 years), respectively. The median interval between mRVOh and the first and second MRIs was 0.58 months (IQR 0.33–6.54 months) and 55.69 months (IQR 38.30–85.43 months), respectively. Comparative analysis of the MRI results revealed significant increases in the mean RV end-diastolic and end-systolic volume indices (P<0.001 for both), mean LV end-diastolic volume index (P=0.040), and both the mean RV and LV stroke volume indices (P<0.001 and P=0.039, respectively), as well as the mean RV cardiac index (P<0.001). No significant differences were found in the PR and ejection fraction of either ventricle. A summary of comparative MRI data, including changes in RV volume indices and cardiac index, is provided in Table 3 and Figure 4.
Table 3.
Comparison of Data From the First and Second Postoperative MRI Scans
| |
Postoperative MRI |
Estimate |
SE |
P value |
| First (n=18) |
Second (n=18) |
| RV ejection fraction (%) |
|
|
−1.883 |
2.916 |
0.528 |
| Mean±SD |
54.99±13.7 |
51.97±6.67 |
|
|
|
| Median [IQR] |
53.4 [47.5–63.1] |
51.0 [48.0–57.4] |
|
|
|
| RV end-diastolic volume index (mL/m2) |
|
|
34.262 |
5.890 |
<0.001 |
| Mean±SD |
56.01±22.28 |
90.68±32.75 |
|
|
|
| Median [IQR] |
53.3 [42.1–70.4] |
87.85 [72–115.1] |
|
|
|
| RV end-systolic volume index (mL/m2) |
|
|
20.140 |
3.357 |
<0.001 |
| Mean±SD |
23.73±14.16 |
44.53±19.33 |
|
|
|
| Median [IQR] |
20.2 [14.3–36.2] |
42.45 [30.9–60.4] |
|
|
|
| RV stroke volume index (mL/m2) |
|
|
16.828 |
3.014 |
<0.001 |
| Mean±SD |
29.71±10.67 |
46.14±15.25 |
|
|
|
| Median [IQR] |
31.2 [22.4–36.7] |
43.35 [38.0–58.2] |
|
|
|
| RV cardiac index (L/min/m2) |
|
|
0.988 |
0.224 |
<0.001 |
| Mean±SD |
3.09±0.93 |
4.05±1.09 |
|
|
|
| Median [IQR] |
3.1 [2.6–3.7] |
4.15 [3.2–4.7] |
|
|
|
| LV ejection fraction (%) |
|
|
−1.061 |
1.608 |
0.519 |
| Mean±SD |
61.82±6.88 |
59.92±4.22 |
|
|
|
| Median [IQR] |
60.8 [57.6–63.6] |
60.4 [56.5–62.9] |
|
|
|
| LV end-diastolic volume index (mL/m2) |
|
|
7.911 |
3.533 |
0.040 |
| Mean±SD |
67.69±15.56 |
76.75±15.46 |
|
|
|
| Median [IQR] |
69.6 [54.3–79.0] |
74.6 [67.3–84.2] |
|
|
|
| LV end-systolic volume index (mL/m2) |
|
|
0.010 |
0.055 |
0.862 |
| Mean±SD |
28.11±7.74 |
30.9±7.63 |
|
|
|
| Median [IQR] |
28.3 [23.1–34.0] |
30.15 [25.2–36.5] |
|
|
|
| LV stroke volume index (mL/m2) |
|
|
4.628 |
2.064 |
0.039 |
| Mean±SD |
41.42±8.66 |
46.34±9.27 |
|
|
|
| Median [IQR] |
42.2 [35.4–47.3] |
44.3 [40.9–53.8] |
|
|
|
| LV cardiac index (L/min/m2) |
|
|
−0.465 |
0.207 |
0.039 |
| Mean±SD |
4.45±1.03 |
4.04±0.62 |
|
|
|
| Median [IQR] |
4.4 [3.5–5.2] |
4.0 [3.5–4.4] |
|
|
|
| Pulmonary regurgitation fraction (%) |
|
|
4.011 |
2.192 |
0.086 |
| Mean±SD |
10.34±12.47 |
14.93±15.47 |
|
|
|
| Median [IQR] |
7.0 [0.52–14.4] |
12.85 [1.3–23.1] |
|
|
|
IQR, interquartile range; LV, left ventricle; MRI, magnetic resonance imaging; RV, right ventricle; SE, standard error.
Predictors of Reoperation, BVR, and Changes in TV, PV, and Cardiac Volume and Function
Predictors of clinical events after mRVOh, such as reoperation and BVR, along with TV and PV annulus size parameters based on echocardiographic findings, are summarized in Supplementary Table 1. Predictors for parameters of cardiac volume and function derived from MRI findings are summarized in Supplementary Table 2.
Discussion
This study has 2 main findings. First, the mean z-scores for the TV and PV annulus increased significantly after mRVOh, showing proportionate growth trends towards the normal range. Second, sequential postoperative MRI studies showed significant increases in RV volume indices and cardiac index while preserving the functions of both ventricles.
Consistent with our prior research,14,15 the TV annulus size increased after mRVOh, achieving catch-up growth trends towards normal ranges rather than merely remaining as disproportionate increases with smaller sizes. Furthermore, the PV annulus size increased after mRVOh. These findings indicate a significant potential for sustainable RV growth. In addition to the echocardiographic study, we performed cardiac MRI to quantitatively assess RV volume and function after mRVOh. Our findings, which suggest catch-up growth trends between postoperative MRI intervals coupled with similar echocardiographic results, may indicate that the mRVOh procedure can promote sustainable optimal RV growth with proportionate TV growth. This contrasts with previous studies reporting suboptimal RV growth with unsatisfactory, disproportionate TV growth.2–4,6,10,13 Even considering that normal volumetric ranges increase with age,20,21 the volume indices of both ventricles after mRVOh remained within the normal ranges throughout follow-up as the children grew. Furthermore, mRVOh may have beneficial effects on the LV.
Rationale for Introducing the RVOh Procedure
PA-IVS encompasses a broad spectrum of congenital heart diseases characterized by varying degrees of RV and TV hypoplasia and coronary artery anomalies.1,2,6–8,12,19 There is no consensus on the optimal surgical strategy to promote RV and TV growth in PA-IVS patients with borderline RV hypoplasia. Treatment strategies for PA-IVS vary across centers due to its inherent heterogeneity, resulting in diverse strategies and outcomes.2,4–13 Previous studies on treatment strategies for PA-IVS, including conventional RVOh, have reported disproportionate TV growth in patients, including those who underwent BVR, despite increases in RV size.2–4,6,10,13 Nonetheless, we believe that timely RV decompression with adequate antegrade pulmonary blood flow through the RV can promote the growth of the RV, TV, and PV, thereby contributing to favorable outcomes and eventual BVR in selected patients with borderline RV hypoplasia without RVDCC. In addition, resection of endocardial fibrotic tissue, which may restrict LV growth, has been reported to improve LV size in patients with borderline-sized LVs.22–24 Nonetheless, reports on endocardial fibrotic tissue and its resection in the RV are limited. However, we have observed hypertrophied RV muscle bundles in addition to endocardial fibrotic tissues in the RV cavity of patients with PA-IVS. Therefore, we postulated that resection of hypertrophied RV muscles in addition to endocardial fibrotic tissue resection could maximize RV growth through earlier RV optimization. In addition, RV volume overload following TR and PR can increase RV size; however, this may represent pathological RV enlargement rather than healthy and true RV growth. To promote healthy RV growth while preserving RV function in the long term, we aimed to minimize these factors to prevent persistent TR and PR. In our cohorts, BPV was attempted in all patients for RV decompression before mRVOh. As a result, most patients (17/23; 73.9%) underwent BPV for RV decompression before mRVOh, except in several patients where BPV failed. Even considering the potential PV leaflet injury caused by BPV, the PR remained within acceptable ranges without appreciable PS after pulmonary valvotomy. Despite the significant increase in PR based on echocardiographic findings, it is of note that the rate of significant (≥moderate) PR remained unchanged when comparing postoperative PR with preoperative PR. Therefore, achieving acceptable outcomes through consistent efforts to minimize PR is critical. Furthermore, considering the dysplastic and hypoplastic characteristics of the TV in patients with PA-IVS, we aimed to minimize TR without causing TS using various TV repair techniques.16 Consequently, we performed mRVOh in patients with PA-IVS without RVDCC and achieved favorable outcomes.14,15 Subsequently, we conducted further studies to evaluate the outcomes of mRVOh as a comprehensive management strategy for PA-IVS with extended follow-up.
Differences From and Potential Advantages Over the Prior RVOh Procedure
Our current strategy for managing PA-IVS differs from that of other centers in several key aspects. Unlike staged repairs, where initial palliative surgery for RV decompression is routinely performed and followed by definitive repair, we performed RVOh as a surgical one-stage primary repair after BPV, aiming to optimize RV growth earlier and more effectively. Accordingly, CPB is essential for this procedure and may provide several potential benefits.
Our CPB-based approach can enable a more precise pulmonary valvotomy under direct vision in a well-exposed field, which may allow for effective PS relief while minimizing potential PR. This approach may reduce the need for future PV-related interventions resulting from potential PV issues and provide long-term benefits in preserving native PV and RV function.5,25 Moreover, if additional interventions such as BPV are required after surgery, precise pulmonary valvotomy could contribute to minimizing the risk of PV leaflet injuries following these interventions.
Performing RVSM and fibrotic endocardial tissue resection in conjunction with pulmonary valvotomy during the initial procedure may ensure effective RV decompression, as well as earlier and more definitive improvements in RV compliance and diastolic function, compared with initial palliation, such as pulmonary valvotomy and/or BTS, alone. Given that studies on conventional strategies with staged repairs showed disproportionate TV growth relative to proportionate RV growth, our procedure may achieve more effective RV growth through early RV optimization.2–4,6,10,13 Unlike traditional approaches that leave hypertrophied RV muscles and endocardial fibrotic tissues during initial palliation, concurrent resection of these tissues in the RV cavity may facilitate more effective RV growth. This hypothesis is supported by our previous study demonstrating that the TV annulus z-score significantly increased after mRVOh compared with BPV alone prior to mRVOh.15 In addition, this procedure may help prevent dynamic RVOT obstruction caused by hypertrophied RV muscles, particularly the infundibular muscles. Therefore, mRVOh may help maintain a more stable antegrade pulmonary blood flow through the PV.
Successful one-stage primary repair may offer additional benefits, such as avoiding RV ventriculotomy, which can impair RV function, and reducing the need for further surgery and inter-stage mortalities associated with BTS.5,26 Conversely, the inability to perform RV decompression through the CPB-based method in patients with RVDCC, due to the risk of coronary ischemia, limits the indication for this procedure to those without RVDCC. Finally, with technological advances in reducing CPB-associated risks in neonates and young infants, this approach is expected to become more widely adopted as a safe method.
Modified RVOh as a Viable Option for Effective RV Growth
Several factors may contribute to RV size increase, including regression of RV hypertrophy due to reduced afterload, improved RV compliance, decreased atrial right-to-left shunting, impact of hypertrophied RV muscle resection, and RV volume overload from varying degrees of TR and PR.4 RV enlargement due to PR and TR indicates pathological remodeling due to volume overload, which is unlikely to sustain long-term RV function. This pathological enlargement differs from ideal and true RV growth. Previous studies have shown that RV volume, as measured by indices such as the RV end-diastolic volume index, increases, whereas TV growth remains insufficient due to the disproportionate increase in TV size.2–4,6,10,13 These findings may also be influenced by the inherent characteristics of a hypoplastic and dysplastic TV, despite considering secondary TV annulus enlargement following pathological RV enlargement.8 This should be considered when evaluating true RV growth in relation to TV growth. In addition, although long-term follow-up studies seldom address the function of the RV, TV, and PV, these factors should be considered in long-term evaluations. Consequently, early RV optimization following mRVOh as the initial procedure for RV decompression may provide several advantages over conventional strategies.
Our findings suggest that, in addition to the sustained increase in RV volume, the TV had grown proportionately while remaining within the normal range. In addition, surgical RV tripartition through RVSM may yield promising results, even in bipartite RV, which may have a high likelihood of a hypoplastic RV with lower TV z-scores.3,4,19 This finding is further supported by our results, which showed that a bipartite RV did not significantly increase the risk of adverse events, including BVR failure, reoperation, or issues with the RV, TV, and PV. In patients with a bipartite RV with underdeveloped trabecular portions, we performed maximum resection of the hypertrophied muscles and endocardial fibrotic tissue toward the apex, aiming for surgical tripartition to promote effective RV growth (Figure 5). In addition, in our cohorts, 14 of 23 (60.9%) patients had a bipartite RV, reflecting our focus on patients with borderline RV hypoplasia. Therefore, we cautiously suggest that exploring the RV growth potential via mRVOh before opting for SVP may be worthwhile, even in patients with smaller RV and TV (TV annulus z-score <−4.0). Significantly, this strategy could broaden future options, including one-and-a-half ventricle repair or BVR, beyond merely SVP. Furthermore, depending on the patient’s follow-up status, redo mRVOh may offer further opportunities for enhanced RV growth. Although one-and-a-half ventricle repair remains a valid option in selected patients, mRVOh may facilitate more balanced RV remodeling while offering physiological advantages. If sustained RV growth can be achieved, the mRVOh strategy may help preserve long-term surgical flexibility and potentially expand durable circulatory options beyond SVP. Notably, no ventricular arrhythmias or signs of significant RV dysfunction were observed during follow-up. This may support the effectiveness of our strategy to preserve RV function, including the avoidance of RV ventriculotomy and tailored TV and PV repairs to optimize RV growth.
Study Limitations
This study has several limitations. First, this study was limited by its non-randomized, retrospective, single-center design and small sample size. Surgical procedures were performed on selected patients without RVDCC, with potential selection bias, because RVDCC typically presents in patients with more severe RV hypoplasia.2,8,19 Comparing groups that underwent only transcatheter interventions or other definitive surgical repairs with the mRVOh group was challenging due to notable inherent differences in baseline characteristics, such as TV annulus size and RV morphology. Notably, because mRVOh was consistently applied to all patients with borderline RV hypoplasia during the study period, no comparable patients with similar anatomical characteristics were managed using alternative strategies. As a result, the absence of an anatomically matched control group treated with other approaches may limit the ability to perform meaningful comparisons and to assess the relative efficacy of mRVOh. Hence, given the wide range of morphologic variability inherent to the broad spectrum of PA-IVS, the findings of this study should be interpreted with caution. Second, the heterogeneity of the study population complicated the control of baseline characteristics and perioperative data due to varying clinical statuses, in addition to the small sample size, despite statistical adjustments. Thus, differences in pre-mRVOh treatment likely affected the outcomes. Furthermore, considering the long follow-up period, advances in surgical techniques and perioperative management over time have limited the generalizability of the findings. Third, most participants did not undergo preoperative MRIs, and the timing of MRIs and the intervals between the first and second postoperative MRI scans during the follow-up period were inconsistent. Despite adjusting for these factors in the statistical analysis, some limitations remained in the evaluation of the results. Fourth, cardiac MRI, highly valued for its well-established reproducibility, was adopted as the primary imaging modality, alongside echocardiography, to quantify RV volume and function. All measurements were obtained according to a standardized, consensus-based protocol to further mitigate potential interobserver variability; nevertheless, such variability cannot be completely eliminated and should be taken into account when interpreting the results. Consequently, long-term studies with larger sample sizes and including assessments of cardiac reserve function, exercise capacity, and reproducibility across multiple centers are warranted.
Conclusions
We observed effective RV growth with proportionate TV annulus growth within the normal range after mRVOh in patients with PA-IVS without RVDCC. Therefore, mRVOh may be a viable and feasible option for facilitating both sustainable RV and TV growth in selected patients with PA-IVS. The procedure may promote RV growth while preserving both RV and TV functions, potentially leading to BVR.
Sources of Funding
This study did not receive any specific funding.
Disclosures
The authors declare that there are no conflicts of interest.
IRB Information
This study was approved by the Seoul National University Hospital Institutional Review Board (Approval no.: H-2402-100-1512; date of approval, March 8, 2024) and adhered to the Declaration of Helsinki.
Data Availability
The deidentified participant data will be shared on a request basis. Please contact the corresponding author directly to request data sharing.
Supplementary Files
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-25-0177
References
- 1.
Joshi SV, Brawn WJ, Mee RB. Pulmonary atresia with intact ventricular septum. J Thorac Cardiovasc Surg 1986; 91: 192–199.
- 2.
Hanley FL, Sade RM, Blackstone EH, Kirklin JW, Freedom RM, Nanda NC. Outcomes in neonatal pulmonary atresia with intact ventricular septum: A multiinstitutional study. J Thorac Cardiovasc Surg 1993; 105: 406–423, 424–427; discussion 423–424.
- 3.
Pawade A, Capuani A, Penny DJ, Karl TR, Mee RB. Pulmonary atresia with intact ventricular septum: Surgical management based on right ventricular infundibulum. J Card Surg 1993; 8: 371–383.
- 4.
Sano S, Ishino K, Kawada M, Fujisawa E, Kamada M, Ohtsuki S. Staged biventricular repair of pulmonary atresia or stenosis with intact ventricular septum. Ann Thorac Surg 2000; 70: 1501–1506.
- 5.
Yoshimura N, Yamaguchi M, Ohashi H, Oshima Y, Oka S, Yoshida M, et al. Pulmonary atresia with intact ventricular septum: Strategy based on right ventricular morphology. J Thorac Cardiovasc Surg 2003; 126: 1417–1426.
- 6.
Bryant R 3rd, Nowicki ER, Mee RB, Rajeswaran J, Duncan BW, Rosenthal GL, et al. Success and limitations of right ventricular sinus myectomy for pulmonary atresia with intact ventricular septum. J Thorac Cardiovasc Surg 2008; 136: 735–742, 742.e1–e2.
- 7.
LaPar DJ, Bacha E. Pulmonary atresia with intact ventricular septum with borderline tricuspid valve: How small is too small. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2019; 22: 27–31.
- 8.
Ashburn DA, Blackstone EH, Wells WJ, Jonas RA, Pigula FA, Mannin PB, et al. Determinants of mortality and type of repair in neonates with pulmonary atresia and intact ventricular septum. J Thorac Cardiovasc Surg 2004; 127: 1000–1007.
- 9.
Odim J, Laks H, Plunkett MD, Tung TC. Successful management of patients with pulmonary atresia with intact ventricular septum using a three tier grading system for right ventricular hypoplasia. Ann Thorac Surg 2006; 81: 678–684.
- 10.
Huang SC, Ishino K, Kasahara S, Yoshizumi K, Kotani Y, Sano S. The potential of disproportionate growth of tricuspid valve after decompression of the right ventricle in patients with pulmonary atresia and intact ventricular septa. J Thorac Cardiovasc Surg 2009; 138: 1160–1166.
- 11.
Liava’a M, Brooks P, Konstantinov I, Brizard C, d’Udekem Y. Changing trends in the management of pulmonary atresia with intact ventricular septum: The Melbourne experience. Eur J Cardiothorac Surg 2011; 40: 1406–1411.
- 12.
Zheng J, Gao B, Zhu Z, Shi G, Xu Z, Liu J, et al. Surgical results for pulmonary atresia with intact ventricular septum: A single-centre 15-year experience and medium-term follow-up. Eur J Cardiothorac Surg 2016; 50: 1083–1088.
- 13.
Kotani Y, Kasahara S, Fujii Y, Eitoku T, Baba K, Otsuki S, et al. A staged decompression of right ventricle allows growth of right ventricle and subsequent biventricular repair in patients with pulmonary atresia and intact ventricular septum. Eur J Cardiothorac Surg 2016; 50: 298–303.
- 14.
Cho S, Kwak JG, Kim WH. Right ventricular sinus myectomy to facilitate right ventricle growth. Oper Tech Thorac Cardiovasc Surg 2022; 27: 206–217.
- 15.
Kwak JG, Kim ER, Yun T, Cho S, Lee CH, Kim WH. Role of surgery on growth of tricuspid valve in pulmonary atresia with intact ventricular septum: Mid-term results of modified right-ventricular overhauling procedure. Congenital Heart Disease 2023; 18: 325–336.
- 16.
Lee JH, Kim WH, Baek SM, Cho SH, Kwon HW, Cho S, et al. Role of tricuspid valve repair in pulmonary atresia with intact ventricular septum: Fresh autologous pericardium as a viable option. Interdiscip Cardiovasc Thorac Surg 2025; 40: ivaf062.
- 17.
Pettersen MD, Du W, Skeens ME, Humes RA. Regression equations for calculation of z scores of cardiac structures in a large cohort of healthy infants, children, and adolescents: An echocardiographic study. J Am Soc Echocardiogr 2008; 21: 922–934.
- 18.
Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD, Levine RA, et al. Recommendations for evaluation of the severity of native valvular regurgitation with two-dimensional and Doppler echocardiography. J Am Soc Echocardiogr 2003; 16: 777–802.
- 19.
Daubeney PE, Delany DJ, Anderson RH, Sandor GG, Slavik Z, Keeton BR, et al. Pulmonary atresia with intact ventricular septum: Range of morphology in a population-based study. J Am Coll Cardiol 2002; 39: 1670–1679.
- 20.
Kawel-Boehm N, Hetzel SJ, Ambale-Venkatesh B, Captur G, Francois CJ, Jerosch-Herold M, et al. Reference ranges (“normal values”) for cardiovascular magnetic resonance (CMR) in adults and children: 2020 update. J Cardiovasc Magn Reson 2020; 22: 87.
- 21.
van der Ven JPG, Sadighy Z, Valsangiacomo Buechel ER, Sarikouch S, Robbers-Visser D, Kellenberger CJ, et al. Multicentre reference values for cardiac magnetic resonance imaging derived ventricular size and function for children aged 0–18 years. Eur Heart J Cardiovasc Imaging 2020; 21: 102–113.
- 22.
Emani SM, Bacha EA, McElhinney DB, Marx GR, Tworetzky W, Pigula FA, et al. Primary left ventricular rehabilitation is effective in maintaining two-ventricle physiology in the borderline left heart. J Thorac Cardiovasc Surg 2009; 138: 1276–1282.
- 23.
Emani SM, McElhinney DB, Tworetzky W, Myers PO, Schroeder B, Zurakowski D, et al. Staged left ventricular recruitment after single-ventricle palliation in patients with borderline left heart hypoplasia. J Am Coll Cardiol 2012; 60: 1966–1974.
- 24.
Gierlinger G, Emani SM. Endocardial fibroelastosis resection: When it works and when it does not. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2024; 27: 19–24.
- 25.
Romeih S, Groenink M, van der Plas MN, Spijkerboer AM, Hazekamp MG, Luijnenburg S, et al. Effect of age on exercise capacity and cardiac reserve in patients with pulmonary atresia with intact ventricular septum after biventricular repair. Eur J Cardiothorac Surg 2012; 42: 50–55.
- 26.
Fenton KN, Pigula FA, Gandhi SK, Russo L, Duncan KF. Interim mortality in pulmonary atresia with intact ventricular septum. Ann Thorac Surg 2004; 78: 1994–1998.
