Improved accuracy in flow mapping of congenital heart disease using stationary phantom technique

Abstract

Background

Flow mapping by cardiovascular magnetic resonance has become the gold standard for non-invasively defining cardiac output (CO), shunt flow and regurgitation. Previous reports have highlighted the presence of inherent errors in flow mapping that are improved with the use of a stationary phantom control. To our knowledge, these studies have only been performed in healthy volunteers.

Results

We analyzed the variation in flow measurements made with and without stationary phantom correction in 31 patients with congenital heart disease. Variation in stroke volume (SV) measurements was seen in all vessels across all patient groups. The variation was largest when analyzing the right ventricular outflow tract (RVOT), with a range of absolute differences in SV from 0.2 to 70 ml and in CO from 0.02 to 4.8 L/min. In patients with repaired Tetrology of Fallot (ToF), the average ratio of pulmonary to systemic blood flow (Qp:Qs) was 1.18 without and 1.02 with phantom correction. Without performing phantom correction, 23% of the repaired ToF patients were classified as having a residual shunt as compared to 0% when flow mapping was performed with phantom correction. Similarly, in patients with known atrial level shunting (ASD/PAPVR) 20% of patients had no shunt when flow mapping was performed without phantom correction as compared to 0% with phantom correction. In patients with bicuspid aortic valves (BAV), the differences in the regurgitant fraction between measuring flow with and without phantom correction ranged from 0 to 30%, while the regurgitant fraction in the RVOT of ToF patients varied by as much as 31%.

Conclusion

The impact of inherent errors in CMR flow mapping should not be underestimated. While the variation across a population may not display a significant trend, for any individual patient it can be quite large. Failure to correct for such variation can lead to clinically significant misinterpretation of flow data. The use of the stationary phantom correction technique appears to improve accuracy both in normal patients as well as those with congenital heart disease.