Cardiac MRI protocol

For all participants, CMR imaging was performed on a 3 T Magnetom Prisma (Siemens Healthcare, Erlangen, Germany) using a combination of an 18-element body matrix coil and 12 elements of the 32-element spine coil. Myocardial tissue alterations were assessed in line with consensus guidelines (Supplementary Table 1 and eAppendix 2) by measuring the longitudinal relaxation time constant of the myocardium (native myocardial T1 time) using a Modified Look-Locker Inversion recovery (MOLLI) sequence, 5 s(3 s)3 s variant in a mid-ventricular short axis slice during breath-hold in end-expiration. The following imaging parameters were used for this electrocardiogram gated single-shot balance steady state free precession sequence: slice thickness 8 mm; flip angle 35°; echo time 1.12 ms; repetition time 280.56 ms; matrix 144 × 256; pixel size 1.4 × 1.4 mm; minimum TI 100 ms; TI increment 80 ms; acquisition window 167 ms; GRAPPA 2 parallel imaging acceleration factor. Circle Cardiovascular Imaging (CVI), Calgary, Canada, version 5.6.3 T1 mapping software was utilised with a Siemens recommended correction factor of 1.035 applied. Each slice was divided into 6 segments as per the American Heart Association model (Fig. ​(Fig.11 and eAppendix 2). An epicardial and endocardial erosion offset of 10% was applied to the contours to ensure only myocardium was included. After visual inspection of all segments and exclusion of those segments with evidence of artefact (e.g., owing to unintended motion effects), a mean T1 relaxation time for the remaining segments of each slice was calculated.

Red and green lines, respectively, represent the endocardial and epicardial borders. Each slice was divided into 6 segments with myocardial T1 time calculated for each segment. Segments with evidence of artefact were excluded from analysis, with the native myocardial T1 time for that participant then calculated as the mean T1 time of the remaining segments

A standard clinical protocol for assessing biventricular function and volumes was followed according to published international guidelines (Supplementary Table 1 and eAppendix 2) and blind to diagnostic group. Volumetric analysis of the cine images was performed using CMRtools (Cardiovascular Imaging Solutions, London, UK) with measurements obtained from the short-axis stack (Fig. ​(Fig.2a)2a) using valve tracking on two corresponding long-axis cines (Fig. 2b–e). Papillary muscle and trabeculations were included in the mass measurement, using a semi-automated signal intensity-based thresholding technique. End-systolic and end-diastolic volumes were calculated thus: as appropriate, in systole or diastole, areas within manually traced endocardial borders (Fig. ​(Fig.2a)2a) were calculated for each short axis cine, and multiplied by slice thickness to provide a slice volume. Slice volumes are measured along the ventricle from apex to the level of the mitral valve (Fig. 2b–e) and summed to calculate overall ventricular volume. Stroke volume and ejection fractions are calculated indirectly from end diastolic and end systolic volumes. Ventricular mass is quantified by measuring the area between endocardial and epicardial borders for sequential short axis cines (Fig. ​(Fig.2a),2a), multiplying by the slice thickness and summing the volume of each slice. Total myocardial volume is multiplied by myocardial density (1.05 g/ml) to provide a measure of myocardial mass. Volumes and mass were indexed to body surface area calculated using the Mosteller formula. Indexed volumetric data were: left ventricular (LV) mass; LV and right ventricular (RV) end-diastolic volumes; LV and RV end-systolic volumes; and LV and RV stroke volumes.

a Left ventricular short axis view with manually traced endocardial and epicardial borders. Semi-automated, signal intensity-based thresholding technique is used to include papillary muscles in the blood pool. b, c Mitral valve tracking in diastole and systole in the vertical long axis. d, e Mitral valve tracking in diastole and systole via a four-chamber view of the heart. Red line depicts level of mitral valve