Unroofed coronary sinus newly diagnosed in adult patients after corrected congenital heart disease

These two patients illustrate different clinical presentations of an UCS as a result of differences between the size of the defect and the degree of left-to-right shunt. Symptoms associated with UCS may vary from none at all, to nonspecific complaints of exertional dyspnoea, to right-sided heart failure due to chronic volume overload as in case B [5, 6]. Snijder et al. described a patient very similar to case B who suffered anginal complaints [7].

These two patients were previously diagnosed with congenital heart disease which had been corrected in childhood. Still the UCS was not diagnosed at that time, demonstrating the difficulty of the diagnosis in a time when imaging modalities were not so advanced as they are nowadays. Current imaging modalities offer the possibility to image the 3D anatomy of the entire heart, which can help in the diagnosis of rare anomalies such as UCS.

In both cases, the diagnostic process was challenging due to concomitant abnormalities. For instance, if the RV dilatation in case A had been considered a consequence of the history of pulmonary regurgitation, for which he had undergone PVR in 2004, then additional diagnostic work-up could have been deemed unnecessary. Similarly, if the PAPVC of the right upper lobe had been considered the cause of RV dilatation despite its small size, then it is unlikely that the UCS would have been discovered during additional tests. It should also be considered that in case A, the size of the left-right shunt was probably limited as a consequence of elevated RV systolic pressure (RVSP) due to progressive pulmonary stenosis. In case B, if the RV dilatation and dysfunction had been ascribed to chronic pacing and tricuspid regurgitation due to the pacemaker lead, then additional diagnostic work-up would not have been performed.

At present, none of the four diagnoses in patient A, i.e. UCS type II, PAPVC of the right upper lobe, moderate pulmonary valve stenosis and a pseudo-aneurysm of the distal anastomosis of the homograft form an indication for intervention. MRI flow measurements across the aorta and pulmonary artery showed that the shunt caused by the UCS and the PAPVC is mild as a consequence of the increased RVSP due to the pulmonary stenosis. The fact that the left-to-right shunt at atrial level is limited is also supported by the fact that the terminal portion of the coronary sinus between left and right atrium was not dilated. The pseudo-aneurysm at the distal anastomosis of the homograft is also not an indication for operation in itself since the risk of progression or rupture is small in the low pressure pulmonary circulation. We decided to postpone surgery until the progressive pulmonary stenosis becomes severe enough to justify surgery. According to the ESC guidelines for the management of grown-up congenital heart disease, an operation is indicated for pulmonary valve stenosis when the peak gradient is >64 mmHg or the RV pressure exceeds 60 mmHg in symptomatic patients (class IC). In asymptomatic patients it should be considered when a severe pulmonary valve stenosis is accompanied by other criteria such as progressive RV dilation or RV systolic dysfunction [8]. In general, reoperation in this population of patients with congenital heart disease is not without risk. Zomer et al. demonstrated that the long-term survival for patients with reoperations in adulthood is inversely correlated to the number of surgeries [9].

The symptoms of patient B are caused by the UCS, which has a pathophysiology similar to an ASD, namely a left-to-right shunt resulting in a volume overload for the right ventricle and the risk of developing pulmonary arterial hypertension. In addition, the PLVCS drains aberrantly in the left atrium causing desaturation. Her symptomatology is most likely explained by progressive pulmonary arterial hypertension, which also causes desaturation due to right-left shunting through the coronary sinus during exercise. The treatment of choice is to close this defect if there is no Eisenmenger syndrome. She was treated with bosentan for 3 months and subsequently underwent right heart catheterisation. The pulmonary pressure measured at 40/17 (mean 24) mmHg was well below 2/3 of the systemic arterial pressure of 114/61 (mean 82) mmHg. The Qp:Qs was 1.4 to 1.0, including the right-to-left shunt. The decision was made for surgical repair of the UCS. While inspecting the left atrium, the PLVCS was confirmed to drain just posteriorly of the left atrial appendage. The coronary sinus was largely unroofed, with the exception of the most distal 2.5 cm, which formed a tunnel connecting the left and right atrium. The coronary sinus was reconstructed with a xenopericard patch over a length of 11 cm. The pacemaker lead was removed from the tricuspid septal leaflet and sutured at the anteroseptal commissure. A tricuspid annuloplasty ring (32 mm) was placed to reduce the annular dimension. At the four-month postoperative follow-up she reported marked improvement in her exercise capacity.

The coronary sinus can be visualised by transthoracic echocardiogram, although this can be difficult because it is located deep posteriorly in the left atrioventricular groove. As with the pulmonary veins, when the coronary sinus is draining into the posterior wall of the left atrium this is not well delineated by transthoracic echocardiogram [10]. In the parasternal long-axis view, it is easy to visualise the coronary sinus when it is dilated. However, demonstrating a defect by echocardiography can be difficult because of the limited sonic window. Transoesophageal echocardiogram is more accurate but it still has the limitations as mentioned above. A dilated coronary sinus should raise the suspicion of a UCS, but in patients with a dilated coronary sinus due to PLVCS, the diameter of the coronary sinus itself does not help to differentiate between those with and without UCS. Kim et al. found that in patients with UCS, the mean diameter of the coronary sinus standardised to the patient’s body surface area was 15 ±4 mm/m², which is similar to that of patients with a PLVCS without UCS (15 ± 6 mm/m²) and significantly greater than of patients with normal anatomy (7 ± 2 mm/m²) [11].

Cardiac CT and MRI are useful for the detection of UCS and other associated anomalies [10, 12]. Cardiac CT has a higher spatial resolution than MRI and in addition the capability of multiplanar reconstruction. Therefore, it will result in better anatomic information [10]. It is also useful for imaging of uncooperative patients because of the short scanning time. An important disadvantage of CT is the radiation exposure, although with modern scanners the radiation dose is quite low. MRI is also very useful for the visualisation of deep lying structures, and does not require radiation [10]. In addition, MRI can provide functional information about left and right ventricular volumes and function, as well as valvular function and shunt size. In our opinion, CT and MRI are complementary in the diagnosis of rare congenital cardiac anomalies such as UCS. In general, these imaging modalities should be used in the context of a targeted diagnostic process, as exemplified by these two cases with incompletely explained RV dilation and dysfunction.

For patients awaiting surgical correction or percutaneous interventions for congenital heart disease a comprehensive imaging work-up should be considered to better appreciate the extent of disease. Technical developments over the last decade have resulted in a substantial increase in image quality that results in a better appreciation of anatomy. Furthermore, the decrease in radiation dose and contrast use make computed tomography imaging more accessible.

Treatment of a UCS consists of surgical rerouting of the coronary sinus and possibly the associated PLVCS to the right atrium using pericardium, a xenopericard patch or left atrium wall. Some defects are so small that closing the holes in the roof of the coronary sinus is sufficient. In the case of an associated PLVCS, when a communicating brachiocephalic vein is present, then the PLVCS can be ligated [12]. However, when the PLVCS ends in the left atrium the brachiocephalic venous bridge is usually absent.

In summary, the diagnosis of a UCS should be considered in a patient with an unexplained enlarged right ventricle, hypoxaemia or desaturation. Clinical suspicion should be particularly high in patients with known congenital heart disease, because there is a strong association between UCS and other anomalies. This diagnosis can remain undetected for decades even in patients who have already undergone cardiac surgery. However, clinical suspicion in combination with modern imaging techniques can help to establish this rare diagnosis.