Magnetic resonance imaging (MRI) can provide accurate anatomical measurements of the cardiac ventricles. This study investigated whether a calculated ventricular mass index (VMI) would provide an accurate means of estimating pulmonary artery pressure noninvasively, and compared the results with conventional Doppler echocardiography and invasive measurement.A total of 26 subjects referred for investigation of pulmonary hypertension were studied by MRI and echocardiography within 2 weeks of cardiac catheterisation. The correlations for mean pulmonary artery pressure were as follows: VMI (ratio of right ventricular mass over left ventricular mass) r=0.81; pulmonary artery systolic pressure (echocardiography) r=0.77. The confidence intervals for the VMI were narrower than for echocardiography. Sensitivity and specificity for pulmonary hypertension were 84 and 71% respectively for the VMI compared with 89 and 57% for echocardiography.The calculated ventricular mass index provides an accurate and practical means of estimating pulmonary artery pressure noninvasively in pulmonary hypertension and may provide a more accurate estimate than Doppler echocardiography. This may be because it reflects the right ventricular response to sustained pulmonary hypertension over a long period and is not influenced by short-term physiological variables affecting echocardiography, such as heart rate, posture, hydration status and oxygen supplementation. Pulmonary arterial hypertension (PAHT) is a rare disorder characterised by high pulmonary vascular resistance. Prognosis is related to mean pulmonary artery pressure (MPAP) measured at right heart catheterisation (RHC) [1], and repeated measurements are often necessary to assess disease progression and the response to treatment. The most widely used noninvasive technique, Doppler echocardiography (ECHO) [2] is safe and widely available, but has several limitations. Firstly, it cannot measure MPAP and only provides an estimate of pulmonary artery systolic pressure (PASP). Secondly, it depends upon the presence of detectable tricuspid regurgitation and has a significant failure rate in some patient groups [3]. Finally, measurements are somewhat operatordependent, and influenced by physiological variables such as heart rate, hydration status and posture, limitations that also affect the accepted gold standard method of RHC [4].Magnetic resonance imaging (MRI) is an attractive modality for studying the complex geometry of the right ventricle and pulmonary vasculature since no assumptions need to be made about the shape or location of the structure being studied. It provides three-dimensional anatomical measurements of right ventricular morphology that are unaffected by physiological variables and more likely to be reproducible than dynamic, planar measurements made at ECHO. Furthermore, these anatomical variables assess the right ventricular response to chronic pulmonary vascular disease and may provide a more clinically relevant assessment of disease severity.MRI has been extensi...