Introduction
A cavopulmonary veno-venous window is a very rare direct communication between the superior caval vein and a right upper pulmonary vein. Reference Butts, Crean and Hlavacek1 A recent report described 1 case and reviewed 7 previously published cases. Reference Nitta, Nakashima and Kino2 Some authors have suggested that a cavopulmonary window is an embryologic remnant of systemic to pulmonary veno-venous connections, while others have proposed that it is a type of sinus venous defect. Reference Van Praagh, Carrera, Sanders, Mayer and Van Praagh3–Reference Aramendi, Cubero, Hamzeh and Rey5 The lesion has been shown to be associated with atrial septal defects by some Reference Zwetsch, Wicky, Meuli and Schnyder6 but not others. Reference Nitta, Nakashima and Kino2 Management of a cavopulmonary window has been predominantly by surgery. Reference Butts, Crean and Hlavacek1,Reference Aramendi, Cubero, Hamzeh and Rey5,Reference Haaf, Kadner, Tabbara and Kessel-Schaefer7 In this report, we describe a 69-year-old male patient incidentally discovered to have a cavopulmonary window after a puzzling course of a transvenous pacing catheter into the left ventricle. The window was closed subsequently with the implantation of covered stents in the superior caval vein.
Case report
A 69-year-old male with a history of atrial fibrillation presented with increasing dyspnoea on exertion. He required a pacemaker due to tachycardia–bradycardia syndrome. During percutaneous pacemaker lead placement at another institution, the lead ended up in the left ventricle, confirmed by chest radiograph (Figure 1S (Supplemental)) and transesophageal echocardiography. The lead was removed and the patient had a CT angiography of the chest, which demonstrated a right upper pulmonary vein–superior caval vein window (Figure 1a–c), moderate right cardiac dilatation, and no significant coronary artery disease. The right upper pulmonary vein drained the anterior and medial aspect of the right upper lobe and connected to the posterior wall of the superior caval vein. The rest of the right upper and middle lobe pulmonary veins drained via a common trunk to the left atrium. A transesophageal echocardiogram showed the window (Figure 2S [Supplemental] A, B, C) and a patent foramen ovale. At catheterisation, the pressures were as follows: right atrium, mean 21 mmHg; right ventricle, 53/23 mmHg; pulmonary artery, 46/24, mean of 34 mmHg; pulmonary capillary wedge, mean of 22 mmHg, descending aorta, 106/69 mmHg, and left ventricle, 112/16 mmHg. Saturations were high superior caval vein 65%, inferior caval vein 77%, right ventricle 89%, pulmonary artery 88%, and aorta 99%. The Qp/Qs was 2.8/1.0. Pulmonary vascular resistance was 2.8 Wood units. The coronary arteries had no significant lesions. The superior caval vein measured 17–18 mm proximally and 30–32 mm distally (Figure 2a–c). A 32 mm x 5 cm Endurant II aortic stent graft extension (Medtronic, Minneapolis, MN) was placed into the distal superior caval vein, and a 4.5 cm covered mounted CP stent (B Braun, Bethlehem, PA) was telescoped into the proximal end of the Endurant graft using a 20 mm balloon (Figure 2d–f). Two small right upper pulmonary veins that drained more cranially into the superior caval vein were sacrificed during stent implantation. Post-procedural transesophageal echocardiography showed unobstructed flow from the right upper pulmonary vein to the left atrium and laminar superior caval vein flow (Figure 2S (Supplemental) D, E, F); the mean pulmonary arterial pressure was 20–22 mmHg.
CT angiography images: ( a ) Axial view demonstrating the window between the right upper pulmonary vein and the superior caval vein (asterisk). ( b ) Right and anterior view and ( c ) left and posterior view of reconstructed images showing the veno-venous window (arrows). Follow-up images: ( d ) axial view showing the stent separating the superior caval vein from the right upper pulmonary vein. ( e ) Reconstructed image from a right and anterior view and ( f ) from a posterior and left view. PV = right upper pulmonary vein; S = superior caval vein.

( a ) Angiogram in the superior caval vein—AP view. ( b ) Angiogram in a small right upper lobe pulmonary vein—lateral view. ( c ) Angiogram in a large right upper pulmonary vein—lateral view. After stent implantation: ( d ) AP view angiogram in the superior caval vein. ( e ) Lateral view angiogram in the superior caval vein. ( f ) Right upper pulmonary vein angiogram—AP view. AP = Anteroposterior; PV = right upper pulmonary vein; S = superior caval vein; R = right atrium.

The patient had an exacerbation of back pain a week after the procedure. An echocardiogram showed a small to moderate-sized pericardial effusion. On CT angiography, there was no residual communication between the superior caval vein and right upper pulmonary vein. There was a mild to moderate circumferential pericardial effusion measuring 10–12 mm in diameter, with a density of 30–50 Hounsfield units, suggestive of a haemorrhagic component. The patient was clinically stable, and the haemoglobin level was unchanged. Five months post-stent implantation, an echocardiogram continued to show a moderate pericardial effusion. A CT angiography showed a similar-sized effusion, and the right upper pulmonary vein and superior caval vein remained unobstructed (Figure 1d–f). Six months after stent placement, he underwent an elective pericardial subxiphoid window with the removal of 400 ml of pericardial fluid. Two years later, the patient was clinically stable; an echocardiogram showed normal biventricular size and systolic function and no pericardial effusion.
Discussion
Some authors have proposed that a cavopulmonary window is a residual embryologic veno-venous connection. Reference Butts, Crean and Hlavacek1 Others have suggested that the lesion is a result of the absence of the wall that divides the right upper pulmonary veins from the superior caval vein. Reference Van Praagh, Carrera, Sanders, Mayer and Van Praagh3 The lesion appears to be distinct from partial anomalous pulmonary venous return to the superior caval vein and a sinus venosus atrial septal defect, in which the superior caval vein can drain into the right atrium or override the atrial septum and drain into both atria. Reference Butts, Crean and Hlavacek1 A similar veno-venous window has been described between the right lower pulmonary veins and inferior caval vein, the latter as a variant or perhaps a separate entity from the inferior type of sinus venosus atrial septal defect. Reference Crystal, Al Najashi, Williams, Redington and Anderson8 Recent interventional procedures placing covered stents in the superior caval vein for partial anomalous pulmonary venous return and sinus venosus atrial septal defect suggest that in some patients, the latter may be a larger “cavopulmonary window” than the one demonstrated in our patient.
Most of the reports on cavopulmonary windows have focused on comorbidities, and the repair, which has included surgery and trans-catheter intervention. Reference Nitta, Nakashima and Kino2,Reference Van Praagh, Carrera, Sanders, Mayer and Van Praagh3 Van Praagh et al. reported 45 patients, studied 1986–1992, with sinus venous defects, and suggested that 39 of them had a “cavopulmonary window.” Reference Van Praagh, Carrera, Sanders, Mayer and Van Praagh3 All of them received surgical intervention. Nitta et al. reported 1 case and added 7 from a literature review of cavopulmonary windows in patients ranging in age from newborns to a 75-year-old. Reference Nitta, Nakashima and Kino2 Among the 8, 2 had patent foramen ovale, and 1 had a hypoplastic left heart. Three patients received surgical repair, 2 had unknown repair, 1 refused repair, and 1 received a covered stent. No details of the stent repair case were provided. Reference Nitta, Nakashima and Kino2 Surgical repair has been the most common repair method for cavopulmonary windows. Reference Nitta, Nakashima and Kino2,Reference Aramendi, Cubero, Hamzeh and Rey5
Our study describes covered stent closure of a cavopulmonary window and an associated complication. We speculate that the left-to-right shunt in our patient was very large, despite only the right upper pulmonary vein connecting to the superior caval vein, because of the large size of the window which drained left atrial blood as well into the superior caval vein, as evidenced by the bidirectional pulmonary venous flow on echocardiography (Figure 2S [Supplemental] C). Our patient developed a chronic pericardial effusion, which we speculate was caused by either small perforations from the stent fixation apices, exacerbated by the patient receiving Xarelto or an inflammatory reaction. The use of stents or stent segments without fixation devices may minimise the risk of a pericardial effusion.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951124036138.
Acknowledgements
The authors thank Dr Yash Desai at the Department of Pediatrics, University of Nevada Las Vegas School of Medicine, for his contributions in editing the manuscript.
Financial support
None.
Competing interests
None.
Ethical standard
The authors assert that this work is exempt of approval from the Institutional Review Board (Sunrise FWA # 00001411).
