Abstract
We present a very rare case of right atrial isomerism, double-outlet right ventricle, and incomplete atrioventricular septal defect (intact ventricular septum). In the neonatal period, the right ventricle was compressed by a “blind-ended” left ventricle with mild-to-moderate left atrioventricular valve regurgitation. The regurgitation gradually decreased from mild-to-moderate to mild with body weight gain. The patient underwent systemic-to-pulmonary shunt at three months of age and bilateral bidirectional Glenn at eight months of age. Although the echocardiogram demonstrated that the right ventricle was still compressed by the left ventricle, over time the size of the left ventricle reduced significantly and the left atrioventricular valve regurgitation became trivial.
Introduction
Hypoplastic left heart syndrome (HLHS) with aortic atresia and mitral stenosis is sometimes associated with a compressed right ventricle (RV) caused by a relatively large left ventricle (LV) with high pressure. 1 Pulmonary atresia with intact ventricular septum (IVS) is also sometimes associated with compressed LV caused by RV with high pressure. 2
We encountered a very rare case of right atrial isomerism (RAI) and double-outlet RV (DORV) with IVS. The RV was compressed by a relatively large LV with mild-to-moderate left atrioventricular valve regurgitation (LAVVR).
Case Report
Fetal echocardiography revealed HLHS, aortic atresia, mitral regurgitation, and a small ventricular septal defect (VSD). A boy (weight, 2,620 g) was delivered by cesarean section at gestational week 37. Postnatal echocardiography and multidetector-row computed tomography confirmed RAI, DORV, incomplete atrioventricular septal defect, single atrium, valvular pulmonary stenosis, patent ductus arteriosus (PDA), ductal-associated pulmonary artery coarctation, right aortic arch, bilateral superior vena cava (SVC), and separate hepatic venous drainage. No vessel arose from the LV and the two great arteries originated from the RV. The right SVC, inferior vena cava, and right hepatic vein (HV) drained into the right-sided atrium. Left SVC and left HV drained into the left-sided atrium (Figure 1). In the neonatal period, the RV was compressed by a relatively large LV (Figure 2A) with mild-to-moderate LAVVR (central regurgitation) (Figure 2B). Flow velocity of the LAVVR exceeded 5.0 m/s. There was no ventriculocoronary connection.

Preoperative multidetector-row computed tomography (A: anterior view, B: posterior view). Color-coding is as follows: aorta and right ventricle (dark grey); left ventricle (black); bilateral superior vena cava, interior vena cava, separate hepatic veins, pulmonary artery and pulmonary vein (light grey); part of atrium (translucent). IVC, inferior vena cava; LHV, left hepatic vein; LSVC, left superior vena cava; LV, left ventricle; RHV, right hepatic vein; RSVC, right superior vena cava; RV, right ventricle.

Echocardiography in the neonatal period demonstrated that the right ventricle (RV) was compressed by a relatively large left ventricle (LV) with mild-to-moderate left atrioventricular valve regurgitation (A: end-diastole, B: end-systole). Echocardiography after Glenn demonstrated that the size of the LV reduced significantly (C: end-diastole).
The PDA was maintained by continuous lipoprostaglandin E1 infusion. LAVVR gradually decreased from mild-to-moderate to mild with body weight gain (Figure 3A). At three months of age (weight, 4.96 kg), the patient underwent systemic-to-pulmonary shunt, pulmonary angioplasty with total resection of PDA tissue, direct end-to-end anastomosis using the patient's pulmonary arterial wall, and division of the main pulmonary artery. At seven months of age, cardiac catheterization showed that the systolic pressure of RV and LV was 93 and 140 mm Hg, respectively. The ejection fraction of RV and LV was 54.5% and 20.3%, respectively. The end-diastolic volume of the RV and LV was 223.0% of normal and 39.0% of normal, respectively.

Time-series data of left atrioventricular valve regurgitation (LAVVR) (A), left ventricular end-diastolic area (LVEDA) (cm2) / body surface area (BSA) (m2) and right ventricular end-diastolic area (RVEDA) / BSA (B) were shown. BBDG, bilateral bidirectional Glenn; MBTS, modified Blalock-Taussig shunt.
The patient underwent bilateral bidirectional Glenn with balanced branch pulmonary arteries at eight months of age (weight, 6.48 kg). Echocardiography demonstrated that the RV was still compressed by the LV. However, the size of the LV had reduced significantly (Figure 2C), and LAVVR became trivial. Left ventricle end-diastolic area (EDA; cm2) divided by body surface area (BSA; m2) changed from 15.0 (day of birth) to 8.3 (before shunt), 6.6 (before Glenn), and 5.3 (2 months after Glenn). The time-series data of LAVVR, LVEDA/BSA, and RVEDA/BSA are shown in Figure 3. The patient is currently awaiting a Fontan completion.
Discussion
Hypoplastic left heart syndrome with aortic atresia and mitral stenosis is associated with “blind-ended” LV. Pulmonary atresia with IVS is also associated with “blind-ended” RV. The pressure in such a “blind-ended” ventricle sometimes exceeds systemic pressure. Consequently, it sometimes compresses the other ventricle1,2 and is associated with ventriculocoronary connection. However, other than these two anatomies, a “blind-ended” ventricle is extremely rare. To the best of our knowledge, DORV with IVS in a patient with RAI has not been reported previously.
In the neonatal period, LAVVR was mild-to-moderate with the relatively large LV compressing the RV (Figure 2A and B). The Starnes procedure can be applied in patients with cardiac anomalies other than Ebstein anomaly and severe uncontrollable atrioventricular valve regurgitation.3,4 Therefore, in cases of severe LAVVR and/or expansion of the LV, we considered performing the Starnes procedure to decompress the LV. However, in this patient, the LAVVR gradually improved and the size of the LV reduced gradually (Figure 3). After we confirmed that the Starnes procedure was unnecessary, we decided on the timing of the first palliation.
We surmised that the reason for both the decrease of LV size and improvement of LAVVR was spontaneous VSD closure just before birth (a small VSD had been detected by fetal echocardiography at 36 weeks of gestation). Consequently, LV pressure overload, LV diastolic dysfunction, trans-left atrioventricular valve volume reduction, and LV size reduction would be expected to occur in chronological order.
As Fogel et al have demonstrated, ventricular septum interactions between the RV and the LV greatly influence systemic ventricle performance.5,6 The “blind-ended” LV compressed the systemic RV in this case. However, the size of the LV gradually decreased with body weight gain. When following up on the size of a “blind-ended” ventricle by echocardiography, the ventricular EDA/BSA is one of the best indicators (Figure 3B).
One may argue that a PDA stent instead of a systemic-to-pulmonary shunt would be less invasive. 7 However, the patient had ductal-associated pulmonary artery coarctation. Therefore, we chose pulmonary angioplasty concomitant with a systemic-to-pulmonary shunt. 8 Consequently, we could perform a bilateral bidirectional Glenn with balanced branch pulmonary arteries.
It may be argued that one and a half biventricular repair using a relatively large LV to pump blood from the IVC and HVs (the hemi-Mustard operation) would be better than univentricular circulation. However, in this patient, we abandoned one and a half repair because of the complex atrial and venous configuration (Figure 1) and decreasing LV size. Moreover, the long suture line involved in the hemi-Mustard in a patient with RAI could result in atrial arrythmia.
The inflow volume into the “blind-ended” LV with trivial LAVVR decreases, in proportion to the reduction of the LV size. However, this can lead to thrombosis in the “blind-ended” ventricle. 9 Therefore, in this case, we prescribed continuous administration of both aspirin and ticlopidine after systemic-to-pulmonary shunt, as well as long-term follow-up and surveillance echocardiography on a regular basis.
Footnotes
Authors' Statement
The permission was granted by the parents of the patient to publish this report.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
