3D and 2D Delayed-Enhancement Magnetic Resonance Imaging for Detection of Myocardial Infarction: Preclinical and Clinical Results

Peukert, Daniel; Laule, Michael; Taupitz, Matthias; Kaufels, Nicola; Hamm, Bernd; Dewey, Marc

doi:10.1016/j.acra.2007.03.006

« Previous Next »Academic Radiology
Volume 14, Issue 7 , Pages 788-794, July 2007

3D and 2D Delayed-Enhancement Magnetic Resonance Imaging for Detection of Myocardial Infarction: Preclinical and Clinical Results

Daniel Peukert, MD
- Department of Radiology, Charité, Humboldt Universität zu Berlin, Institut für Radiologie, Charitéplatz 1, 10117 Berlin, Germany
Michael Laule, MD
- Department of Cardiology, Charité, Humboldt Universität zu Berlin, Institut für Radiologie, Charitéplatz 1, 10117 Berlin, Germany.
Matthias Taupitz, MD
- Department of Radiology, Charité, Humboldt Universität zu Berlin, Institut für Radiologie, Charitéplatz 1, 10117 Berlin, Germany
Nicola Kaufels, DVM
- Department of Radiology, Charité, Humboldt Universität zu Berlin, Institut für Radiologie, Charitéplatz 1, 10117 Berlin, Germany
Bernd Hamm, MD
- Department of Radiology, Charité, Humboldt Universität zu Berlin, Institut für Radiologie, Charitéplatz 1, 10117 Berlin, Germany
Marc Dewey, MD
- Department of Radiology, Charité, Humboldt Universität zu Berlin, Institut für Radiologie, Charitéplatz 1, 10117 Berlin, Germany
- Address correspondence to: M.D.

Received 14 December 2006; accepted 10 March 2007.

Rationale and Objectives

The purpose was to verify whether myocardial viability can be detected by a delayed enhancement magnetic resonance imaging (MRI) approach using a rapid three-dimensional inversion-recovery fast low-angle shot (3D IR-FLASH) sequence in a preclinical and clinical setting.

Materials and Methods

Nonreperfused myocardial infarctions were induced in eight minipigs. Both the pigs and 15 patients with suspected myocardial infarction underwent MRI using a rapid 3D IR-FLASH sequence and a two-dimensional IR-FLASH sequence as the reference standard.

Results

In the pigs, a total of 52 segments with myocardial infarction were identified with both sequences and there was good agreement in transmurality of 99.5%. The infarction volume determined with the 3D IR-FLASH in the animal study (2.4 ± 1.5 cm³) showed a good correlation with the histomorphometrically determined volume using triphenyltetrazolium chloride (2.3 ± 1.2 cm³, r = 0.98, P < .001) and the two-dimensional IR-FLASH sequence (2.3 ± 1.4 cm³, r = 0.99, P < .001). Eleven of 15 patients were found to have myocardial infarction in 37 myocardial segments with both sequences and there was a good agreement in transmurality of 98.8%. There was also a good correlation in the clinical study between the 3D and 2D sequences (6.9 ± 6.7 cm³ vs. 6.8 ± 6.5 cm³, r = 0.98, P < .001). In Bland-Altman analysis there was no significant under- or overestimation of the myocardial infarction volume using the 3D IR-FLASH sequence in comparison to the two-dimensional reference standard in both the preclinical and clinical study. The contrast-to-noise ratios were not significantly different between 3D and 2D sequences in the animal (34.7 ± 1.5 vs. 33.8 ± 2.6; P = .51) and clinical study (31.4 ± 12.5 vs. 36.7 ± 11.5; P = .31). The breathhold time for the 3D IR-FLASH sequence in the clinical study (20.4 ± 2.2 s) was significantly shorter than that of the 2D IR-FLASH sequence (190.1 ± 20.8 s, P < .001).