Cardiac Magnetic Resonance Stress Imaging: A Feasibility Study in Healthy Young Volunteers

Cardiac magnetic resonance imaging (MRI) is the clinical gold‐standard imaging tool for evaluation of resting left ventricular morphology and function. Cardiovascular stress testing is known to increase the prognostic value of cardiac imaging, but remains limited to non‐MR modalities due to a lack of MRI‐compatible equipment and/or sophisticated MRI sequencing and post‐processing. As a result, MR stress imaging relies heavily on pharmaceutical agents (e.g. dobutamine, adenosine) which do not fully recapitulate the physiological response to activities of daily living. Here, we present preliminary data on 11 healthy volunteers (7 women, 34 ± 8 years of age, BMI: 24 ± 3 kg/m 2 ), using a commercially available, MR compatible, leg ergometer (Ergospect, Tirol, Austria), with standardized, vendor provided, imaging sequences (Siemens Healthcare, Erlangen, Germany). Imaging was performed using a Siemens 3T Magetom Verio MR system. Cine images of the left ventricular long‐axis (4‐chamber, 2‐chamber) and mid‐ventricular short‐axis were acquired at end‐inspiration, with standard steady‐state free precession imaging (temporal resolution ~17 ms, TE = 1.3 ms, slice thickness = 8 mm, matrix size = 224 x 224, and FOV = 300 mm). To evaluate changes in regional tissue deformation, MR tissue tagging was also performed in the mid‐ventricular short‐axis orientation, at end‐inspiration (tag spacing = 10 – 12 mm, temporal resolution ~24.2 ms, TE = 1.8 ms, slice thickness = 8 mm, matrix size = 224 x 224, and FOV = 330 mm). Subjects were instructed to avoid excessive upper body movement. Exercise resistance was adjusted for each individual (ranging from 50 W to 200 W, depending on fitness level and body habitus), with a target heart rate of 100 beats/min. Image acquisition during exercise was possible in 9 of the 11 subjects. Magneto‐hydrodynamic turbulence prevented EKG‐gating in the remaining subjects, disallowing MR imaging during exercise. In those subjects, image acquisition was performed immediately after repeat bouts of exercise cessation. We successfully increased heart rate from 65 ± 12 bpm to 93 ± 10 bpm (P < 0.001). Left ventricular end‐diastolic volume decreased slightly with exercise (from 127.3 ± 27.9 mL to 119.8 ± 32.5 ml), with a similar reduction in end‐systolic volume (from 43.6 ± 11.1 to 35.0 ± 14.7 ml). As a result, left ventricular stroke volume was maintained with exercise (from 83.7 ± 20.5 to 84.8 ± 22.2 ml), while cardiac output increased by more than 2 L per minute (P < 0.01). Global left ventricular systolic function (i.e. ejection fraction) increased with exercise (from 66 ± 5% to 72 ± 8%); whereas, regional tissue deformation (i.e. left ventricular circumferential strain) did not significantly change (from −21.3 ± 2.0% to −20.9 ± 2.5%). We did however observe a significant increase in peak systolic circumferential strain rate (from −114.4 ± 13.4 %/s to −141.5 ± 25.0 %/s, P < 0.01) and peak early diastolic strain rate (from 172.8 ± 25.9 %/s to 203.7 ± 44.8 %/s, P < 0.05). Taken together, these encouraging data support the feasibility of exercise MRI, and provide basic proof‐of‐concept that such imaging does not require sophisticated image sequencing and/or post‐processing. Next steps will involve application of this work to clinical populations. Support or Funding Information This work was supported by AHA 16SDG27260115 and the Harry S. Moss Heart Trust.