David Capener, Abstract Co-Author: Nothing to Disclose

Andrew James Swift BMedSc, FRCR, Presenter: Nothing to Disclose

Smitha Rajaram FRCR, Abstract Co-Author: Nothing to Disclose

David Kiely MD, Abstract Co-Author: Advisory Board, Eli Lilly and Company Advisory Board, Actelion Ltd Advisory Board, Bayer AG Advisory Board, glaxoSmithKline plc Research funded, Actelion Ltd Research funded, Pfizer Inc Research funded, Bayer AG

Judith Hurdman MD, Abstract Co-Author: Nothing to Disclose

Robin Condliffe MD, Abstract Co-Author: Nothing to Disclose

Charlie Elliot MBBS, Abstract Co-Author: Nothing to Disclose

Jim M. Wild PhD, Abstract Co-Author: Nothing to Disclose

The aim of this study was to compare 2D velocity encoded phase contrast MRI and 2D short axis stack CINE MRI techniques with invasive haemodynamic measurements for the estimation of cardiac output (CO) in patients with suspected pulmonary hypertension.

110 consecutive patients with suspected pulmonary hypertension underwent 2D phase contrast MRI and CINE cardiac MRI at 1.5T. Phase contrast velocity vector measurement analysis was performed using a single slice positioned at the level of the pulmonary artery trunk using a gated 2D velocity encoded phase contrast sequence, from which average blood flow (L/min) was calculated. CO was measured using 2D FIESTA short axis stack CINE cardiac MRI by multiplying stroke volume by heart rate. CO (L/min) was also measured at right heart catheterisation (RHC) using the thermodilution technique for comparison with our cardiac MR indices.

Average blood flow calculated using phase contrast MRI correlated well with CO measured at RHC, r=0.74 (p<0.0001) Figure 1. CO measured using 2D FIESTA short axis stack CINE cardiac MRI showed a weaker correlation with the catheter derived CO, r=0.2 (p=0.036). ROC curve analysis indicated that phase contrast MRI has high sensitivity and specificity for detecting reduced CO (area under the ROC curve=0.89), CO estimated from CINE cardiac MRI had lower sensitivity and specificity (area under the ROC curve=0.60). Bland Altman analysis revealed a mean difference between phase contrast MRI measured CO and RHC derived CO of -1.4 (-3.9 to 1.2), weaker agreement was demonstrated for CO measured by CINE cardiac MRI -2.6 (-7.0 to 1.8).

CO assessed using velocity encoded phase contrast MRI more closely reflects CO measured from invasive RHC than CO estimated from 2D FIESTA short axis stack CINE cardiac MRI. The technique provides a useful complementary tool in the evaluation of patients with suspected PH.

2D velocity encoded phase contrast MRI derived CO demonstrates a close correlation to right heart catheterisation CO and is recommended as part of the MRI study in suspected pulmonary hypertension.

Capener, D, Swift, A, Rajaram, S, Kiely, D, Hurdman, J, Condliffe, R, Elliot, C, Wild, J, 2D Velocity-encoded Phase Contrast MRI Better Reflects Cardiac Output than 2D FIESTA CINE Short Axis Stack Cardiac MRI in Patients with Suspected Pulmonary Hypertension.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11015143.html