Radiologists have long held great expectations for cardiac MRI in the diagnosis of coronary artery disease. Its developers envisioned comprehensive applications encompassing the diagnosis of myocardial infarction, detection and measurement of coronary artery stenoses, and assessment of myocardial viability.

Overcoming MRI's poor temporal resolution to realize that vision has proved difficult, however. Despite some progress, coronary artery MRA is still not ready to challenge x-ray angiography, and although first-pass perfusion MRI may match the accuracy of stress echocardiography and thallium-201 SPECT, it remains difficult to perform.

These shortcomings provided an opening for MRI myocardial viability assessment. Since 1997, studies by Dr. Charles Higgins, Dr. Robert M. Judd, Dr. Raymond Kim, Dr. Christoph Klein, Dr. Joao A. Lima, and Dr. Jorn J. Sandstede and others have shown that late-enhancement MRI accurately predicts whether patients will benefit from revascularization.

Dr. Ernest E. Van der Wall, a professor of cardiovascular radiology at the University of Leiden in the Netherlands, said at the SCMR meeting that the clinical efficacy of delayed-enhancement MRI is no longer in doubt.

"Bright is dead," he said, referring to the hyperenhanced depiction of myocardial infarction on late-enhancement images.

Other studies reported at the SCMR meeting reinforced that conclusion and suggested that late enhancement can play other roles.

Dr. Peter Hunold, a resident in radiology at University Hospital in Essen, Germany, presented a study of 12 CAD patients in which late-enhancement MRI was shown to be more specific and to have a higher positive predictive value than FDG-PET for predicting the success of revascularization. No significant difference in negative predictive values for the two modalities was seen.

Based on the evaluation of 1008 myocardial segments, the sensitivity and specificity of MRI for predicting functional recovery was 95% and 72%, respectively. The FDG-PET results for these measures were 97% and 47%, respectively.

"The specificity of MRI is not optimal, but it is remarkably higher than PET, because MRI is generally less optimistic than PET," Hunold said.

The PPV and NPV for regional myocardial improvement were 66% and 96% for MRI and 52% and 97% for PET.

"Our answer to the question whether MRI should be the new gold standard for myocardial viability would certainly be 'yes,'" he said.

SPECT COMPARISON

In a study of 53 patients with left ventricular dysfunction following MI, Dr. Matthias Regenfus, a cardiovascular radiology researcher at Frederick Alexander University in Erlangen Nurnberg, Germany, found a close correlation between late-enhancement contrast-enhanced MRI and thallium-201 SPECT for evaluation of myocardial viability. The segmental extent of infarction was significantly less pronounced on the SPECT studies, however, and SPECT failed to identify 70 of 352 myocardial segments that hyperenhanced on CE-MRI.

Regenfus credited CE-MRI's relatively high spatial resolution for its superior performance.

"It allows late-enhancement CE-MRI to better depict the transmural extent of myocardial infarction and detect small myocardial necroses that escape detection with SPECT," he said.

Research from Duke University indicates that CE-MRI may be a better way to diagnose silent MI, according to Dr. Han Kim, a cardiovascular MRI research fellow.

The hypothesis that CE-MRI may be more sensitive to silent MI than Q-wave/non-Q-wave ECG, the current diagnostic gold standard, was supported by a study of 100 patients with suspected silent MI. Hyperenhancement of infarction with CE-MRI was observed in 57% of the cases, a rate four times higher than the 14% rate of unrecognized MI detected with the Q-wave test, Kim said.

Follow-up evaluations after an average of 21 months found that the extent of hyperenhancement was the only statistically significant predictor of death (p = 0.002). The risk of death rose when the hyperenhanced infarction involved more than 5% of the left ventricular myocardium.

Several studies at the SCMR conference examined how to distinguish acute from chronic MI with CE-MRI. Dr. Hassan Abdel-Aty, a cardiology researcher at the Franz-Vohard Clinic in Berlin, found that late-enhancement imaging combined with short T1 inversion recovery (STIR) imaging can play that role.

In a study of 119 patients (74 with acute MI, 27 with chronic MI, and 18 healthy controls), late-enhancement MRI was 100% sensitive to infarction but was unable to differentiate acute from chronic disease. STIR, however, revealed transmural hyperintensity among 71 of 74 acute infarcts and no transmurality among the chronic lesions. The sensitivity and specificity of the paired tests were 96% and 100%, respectively.

IMAGING CARDIOMYOPATHY

Late-enhancement CE-MRI may also find a role in the diagnosis of cardiomyopathies. Dr. Marcelo Hadlich, a radiology researcher associated with the D'Or Hospital Network in Rio de Janeiro, Brazil, reported that delayed-enhancement MRI can detect irreversible myocardial disease in patients with myopericarditis.

A preliminary study of five patients found delayed hyperenhancement in 40% of the left ventricular segments examined. Those areas were small, diffusely distributed, and not restricted to any specific coronary territories, Hadlich said. The same hyperenhancement appeared when imaging was repeated three months later, suggesting irreversible myocardial injury.

The work of Dr. James C.C. Moon and colleagues at Royal Brompton Hospital in London, U.K., suggests that cine CMR and CE-MRI myocardial hyperenhancement may diagnose early hypertrophic cardiomyopathy and serve as a marker for clinical risk.

In a study of 30 patients from 13 families with mutations of troponin I, echocardiography found hyperenhancement (LVH+) and exaggerated wall thickness in 15 subjects, and no abnormal enhancement and normal wall thickness among 15 (LVH-) individuals who are either in the early stages or disease-free.

Abnormal cine CMR and abnormal regional hypertrophy were found in 100% of the LVH+ and 27% of the LVH- subjects. Hyperenhancement appeared in 86% of the LVH+ and 20% of the LVH- groups.

The overall extent of hyperenhancement was related to the risk of sudden death and total left ventricular mass, and it was inversely related to left ventricular ejection fracture, Moon said.

Although no age-related pattern was observed for the group as a whole, hyperenhancement increased with age compared with the subjects' family members. This pattern suggests that increasing hyperenhancement patterns over time reflect disease progression, he said.

Chagas disease, also called South American trypanosomiasis, is caused by a blood-borne parasite and can lead to myocardial fibrosis, typically in apical and basal inferolateral left ventricular segments. Dr. Carlos E. Rochitte at the Heart Institute of the University of Sao Paulo Medical School in Brazil found that delayed-enhancement MRI identified fibrosis in 22 of 25 Chagas disease patients and in 31% of the myocardial segments that were observed. More midwall and subepicardial involvement was documented than was considered typical with the disease. The extent of myocardial fibrosis correlated well with left ventricular ejection fraction.

Although more investigative work is needed to refine these applications, late-enhancement CE-MRI is quickly gaining favor among researchers responsible for the creation and development of cardiac MRI.