Classification of diastolic function with phase-contrast cardiac magnetic resonance imaging: validation with
Sebastian J Buss1, Birgit Krautz, Bernhardt Schnackenburg
1Department of Cardiology, Angiology and Pneumology, University of Heidelberg, Im Neuenheimer Feld 410, Heidelberg, 69120, Germany, sebastian.buss@med.uni-heidelberg.de.
Objectives:
To investigate whether cardiac magnetic resonance phase-contrast imaging (PC-CMR) can determine left ventricular (LV) diastolic function in comparison to echocardiography (EC).
Background:
Non-invasive evaluation of diastolic function is important for the diagnostic classification and risk stratification of patients with cardiomyopathies. With EC, diastolic function is classified based on the mitral blood flow, LV myocardial tissue Doppler velocities and pulmonary venous flow. PC-CMR has the potential to measure these parameters and may be an important tool to assess diastolic function in clinical routine.
Methods:
In 36 patients with various cardiovascular diseases and 6 healthy volunteers, we performed single-slice short-axis PC-CMR at the level of the mitral leaflet tip and the inflow of the pulmonary veins to generate EC-comparable mitral E and A waves, septal and lateral e' and a' tissue velocities, and E/A and E/e' ratios. EC was performed after PC-CMR in all patients and six volunteers. Patients were classified into three groups of DD for both techniques. In addition, we evaluated 120 healthy volunteers as controls (3 age groups: 1 = 20-35 years; 2 = 36-50 years; 3 ≥ 51 years) for reference values.
Results:
PC-CMR correlation with EC regarding the relation of mitral E and A velocities was good (r = 0.83, p < 0.001). The correlation for the mean septal and lateral E/e' ratio was high with r = 0.90 (p < 0.001). 40/42 subjects (95 %) were categorized correctly. The mean scan time for PC-CMR was 189 ± 16 s and mean analysis time was 348 ± 95 s. EC image acquisition time was slightly higher (201 ± 37 s, p = n.s.), whereas EC image analysis time was significantly lower (149 ± 23 s, p < 0.001).
Conclusion:
The classification of DD with PC-CMR is feasible and shows good agreement with the widely accepted EC classification of DD. We present a practical approach for the clinically important assessment of DD with PC-CMR, circumventing sophisticated and time-consuming CMR sequences and specially designed software analysis tools.
