Cardiac magnetic resonance imaging predicts cardiac catheter findings for great artery stenosis in children with congenital cardiac disease

Objective: To assess the cardiac catheterisation findings of all children in whom cardiac magnetic resonance imaging found great artery stenosis. Methods: We conducted a retrospective analysis of all 45 consecutive children with congenital cardiac disease who were undergoing cardiac catheterisation for intervention on cardiac magnetic resonance-defined great vessel stenosis, between January, 2006 and August, 2008. Results: Following cardiac magnetic resonance, 60 significant great vessel stenoses were identified and referred to cardiac catheterisation for intervention. All patients were catheterised within a median and interquartile range of 84 and 4-149 days, respectively, of cardiac magnetic resonance. At cardiac catheterisation, the children were aged 11.5 years - with an interquartile range of 3.8-16.9 years - and weighed 34 kilograms - with an interquartile range of 15-56 kilograms. Comparing cardiac magnetic resonance and cardiac catheterisation findings, 53 (88%) findings were concordant and seven were discordant. In six of seven (86%) discordant observations, cardiac magnetic resonance defined moderate-severe great vessel stenosis - involving three branch pulmonary arteries and three aortas. This was not confirmed by cardiac catheterisation, which revealed mild stenoses and haemodynamic gradients insufficient for intervention. In one patient, a mild, proximal right pulmonary artery narrowing was found at cardiac catheterisation, which was not mentioned in the cardiac magnetic resonance report. There was no difference between discordant and concordant groups on the basis of patient age, weight, interval between cardiac magnetic resonance and cardiac catheterisation, or type of lesion. Conclusion: Invasive assessment confirmed cardiac magnetic resonance-diagnosed great vessel stenosis in the majority of this cohort. The predominant discordant finding was lower catherisation gradient than predicted by morphologic and functional cardiac magnetic resonance assessment. Flow volume diversion - for example, unilateral pulmonary artery stenosis - and anaesthetic effects may account for some differences. Prospective refinement of cardiac magnetic resonance and interventional data may further improve the validity of non-invasive imaging thresholds for intervention.