Establishment and Performance of a Magnetic Resonance Cardiac Function Clinic

• Published in: Journal of cardiovascular magnetic resonance Vol. 2; no. 1; pp. 15 - 22
• Main Authors: Bellenger, Nicholas G., Francis, Jane M., Davies, Ceri L., Coats, Andrew J., Pennell, Dudley J.
• Format: Journal Article
• Language: English
• Published: England Informa UK Ltd 01.01.2000
• Subjects: Cardiology Service, Hospital - organization & administration; Echocardiography; Female; Heart Diseases - diagnosis; Humans; Image Processing, Computer-Assisted; Magnetic Resonance Imaging; Male; Middle Aged; Outpatient Clinics, Hospital - organization & administration; Patient Satisfaction; Radionuclide Ventriculography; Reproducibility of Results; Statistics, Nonparametric
• ISSN: 1097-6647; 1532-429X
• DOI: 10.3109/10976640009148669

Our objective was to establish a cardiovascular magnetic resonance (CMR) cardiac function clinic to provide an assessment of cardiac volume, mass, and function in patients with heart failure on the same day as their cardiology outpatient clinic appointment. Sixty-four patients attended the CMR function clinic. The reproducibility, patient acceptability, and time efficiency of the CMR clinic were assessed and compared with radionuclide ventriculography (RNV) and echocardiography (echo). Reports were available in the cardiology outpatient clinic within 2 hr of the CMR appointment time. The reproducibility of volumes, ejection fraction, and mass in this heart failure population was good and comparable with CMR studies in the normal population. CMR was more acceptable to the patients than both RNV and echo (p < 0.05). The total time for CMR was less than that of RNV (42 ± 4 and 61 ± 4 min, respectively; p < 0.001) but more than that of echo (echo, 23 ± 2 min; p < 0.001). Comparison of ejection fractions revealed a correlation between CMR and RNV of 0.7, but Bland-Altman limits of agreement were wide (-10.5% to 18.9%). For CMR versus echo, the correlation was 0.6, and the limits of agreement were wider (-29.9% to 23.3%). The correlation between RNV and echo was 0.2 with wider limits of agreement (-29.8% to 24.9%). In conclusion, CMR can provide a rapid, reproducible, and patient acceptable assessment of cardiac function in heart failure patients, whereas other methods appear to have a wider variance. The high reproducibility of CMR lends itself to the follow-up of clinical progression and the effect of treatment in patients with heart failure.