Flow profiles are important determinants of fluid–vessel wall interactions. The aim of this study was to assess blood flow profiles in the aorta and pulmonary trunk in patients with transposition and different ventriculoarterial connection, and hence different mechanics of the coherent pump.

In all, 29 patients with operated transposition – concordant atrioventricular and discordant ventriculoarterial connection, and no other cardiac malformation – and eight healthy volunteers were assessed with cardiac magnetic resonance imaging: n = 17 patients after atrial redirection, with a morphologic right ventricle acting as systemic pump and a morphologic left ventricle connected to the pulmonary trunk, and n = 12 patients after the arterial switch procedure, with physiologic ventriculoarterial connections. Flow-sensitive four-dimensional velocity-encoded magnetic resonance imaging was used to analyse systolic flow patterns in the aorta and pulmonary trunk, relating to helical flow and vortex formation.

In the aorta, overall helicity was present in healthy volunteers, but it was absent in all patients independent on the operation technique. Partial helices were observed in the ascending aorta of 58% of patients after arterial switch. In the pulmonary trunk, mostly parallel flow was seen in healthy volunteers and in patients after arterial switch, whereas vortex formation was present in 88% of patients after atrial redirection.

Blood flow patterns differ substantially between the groups. In addition to varying mechanics of the coherent pumping ventricles, the absent overall helicity in all patients might be explained by the missing looping of the aorta in transposition.

To assess if flow-sensitive four-dimensional velocity-encoded cine magnetic resonance imaging adds value in diagnosing patients with suspected partial anomalous pulmonary venous drainage.

In six patients with echocardiographically suspected partial anomalous pulmonary venous drainage, anatomy was evaluated using standard magnetic resonance imaging including angiography. Functional analysis included shunt calculations from flow measurements. We used four-dimensional velocity-encoded cine magnetic resonance imaging for visualisation of maldraining pulmonary veins and quantification of flow via the maldraining veins and interatrial communications, if present.

In all patients, the diagnosis of partial anomalous pulmonary venous drainage was confirmed by standard magnetic resonance imaging. Shunt volumes ranged from 1.4:1 to 4.7:1. Drainage sites were the superior caval vein (n = 5) or the vertical vein (n = 1). Multiple maldraining pulmonary veins were found in three patients. Pulmonary arteries and veins could be clearly distinguished by selective visualisation using four-dimensional velocity-encoded cine magnetic resonance imaging. Flow measured individually in maldraining pulmonary veins in six patients and across the interatrial communication in three patients revealed a percentage of the overall shunt volume of 30–100% and 58–70%, respectively.

Selective visualisation of individual vessels and their flow characteristics by four-dimensional velocity-encoded cine magnetic resonance imaging facilitates in distinguishing adjacent pulmonary arteries and veins and thus improves the accurate diagnosis of maldraining pulmonary veins. By detailed quantification of shunt volumes, additional information for planning of treatment strategies is provided. This method adds clinical value and might replace contrast-enhanced magnetic resonance angiography in these patients in the future.