Magnetic resonance imaging in determination of myocardial ischemia and viability: comparison with positron emission tomography and single-photon emission computed tomography in a porcine model.

BACKGROUND

In patients with chronic left ventricular dysfunction, the size of the viable cardiac muscle is correlated with the prognosis and the outcome of myocardial revascularization.

PURPOSE

To evaluate the diagnostic value of various imaging techniques in determination of myocardial ischemia and viability.

MATERIAL AND METHODS

A chronic myocardial ischemia animal model was established, in which 10 pigs underwent magnetic resonance imaging (MRI), positron emission tomography (PET), and single-photon emission computed tomography (201Tl SPECT) before and 1-2 months after modeling. The size of myocardial ischemia and necrosis was judged, and the imaging manifestations were compared with pathologic findings.

RESULTS

Seven of the 10 animals completed all examinations uneventfully. On dobutamine-stressed cine MRI, 10 (8.93%) segments were found to be akinetic. Perfusion was abnormal in 34 (30.35%) segments. Delayed hyperenhancement was observed in 12 (10.71%) segments. PET detected myocardial necrosis in 17 (15.18%) segments, and SPECT detected myocardial necrosis in nine (8.04%) segments. Histological examination with triphenyltetrazolium chloride (TTC) showed pale necrosis in 14 (12.50%) segments. The number of necrotic segments detected by PET was significantly greater than that by contrast-enhanced MRI (chi2 = 5, P = 0.0253, kappa = 0.8028) and cine MRI (chi2 = 7, P = 0.0082, kappa = 0.7079). It was also greater than that by TTC (chi2 = 3, P = 0.0833, kappa = 0.8879), although the difference was statistically insignificant. The number of necrotic segments detected by SPECT was significantly smaller than that by TTC (chi2 = 5, P = 0.0253, kappa = 0.7590), as was the number of necrotic segments detected by cine MRI (chi2 = 4, P = 0.0455, kappa = 0.8100). There was no statistically significant difference in the detection of necrotic segments between contrast-enhanced MRI and TTC (chi2 = 2, P = 0.1573, kappa = 0.9130).

CONCLUSION

Cardiac MRI can determine viable myocardium and clearly delineate the location and degree of myocardial necrosis. PET slightly overestimates the extent of the necrotic myocardium and is unable to distinguish transmural necrosis from subendocardial necrosis.