64-slice cardiac CT proves powerful but not perfect

Sixty-four-slice CT can be a potent noninvasive tool for imaging coronary arteries, but it falls down in the presence of calcifications and in per-segment analysis, according to research presented at the European Congress of Radiology in Vienna in March.

One study looked at the use of 64-slice CT angiography as an alternative to coronary angiography in patients with low to moderate risk of coronary artery stenosis.

"Many patients have equivocal symptoms and an intermediate risk profile, so a reliable method for identification of significant artery stenosis would be of great value," said Dr. Thomas Schlosser of the University Hospital Essen in Germany.

The study, which was performed by the University of Essen and the Cardiovascular Center Bethanien in Frankfurt, involved 177 patients (two-thirds men) with suspected coronary artery disease imaged on a 64-slice system with reconstructed slice thickness of 0.6 mm. Researchers were able to confidently rule out 116 patients for high-grade coronary artery stenosis (greater than 50%), and these patients were spared invasive angiography.

The 61 remaining patients (also two-thirds men) underwent invasive coronary angiography. CTA found 62 cases of significant stenosis on 915 segments. Coronary angiography confirmed 34 true positives and identified 28 false positives. These resulted from severe calcification and from plaque formation that did not produce a significant reduction of the lumen.

"In many cases, CTA depicted plaques in the coronary arteries, which appeared to result in a significant lumen narrowing that could not be confirmed on invasive coronary angiography," Schlosser said. "So one important limitation of multislice CT is the fact that mild or moderate coronary artery stenoses can be overestimated, particularly in the presence of calcification."

The utility of 64-slice CTA proved strong despite this limitation. The technique demonstrated sensitivity of 100%, specificity of almost 97%, a positive predictive value of 54.8%, and a negative predictive value of 100%. The negative predictive value of 100% suggests that patients with angiographically normal coronary arteries can be correctly identified using MSCT, Schlosser said.

"Sixty-four-slice CT reliably detects significant coronary artery stenosis in patients with low to moderate pretest probability of significant coronary artery disease," he said. "CTA appears to be helpful for selecting patients who need to undergo invasive angiography."

In another study, researchers from the Medical University of South Carolina in Charleston examined 64-slice CT in coronary imaging overall, and then by segment and by vessel. They also evaluated performance for stenosis greater than 50% and 70%. Sensitivity and specificity with 64-slice CT were very good on a per-patient basis, indicating that the imaging technique may be used to rule out significant coronary artery disease, according to Dr. Giancarlo Savino of the MUSC radiology department.

In cases of more than 50% stenosis, CTA produced the following results:

- accuracy of 99%;

- sensitivity of 100%;

- specificity of 98%;

- PPV of 98%; and

- NPV of 100%.

In stenosis greater than 70%, values were lower but still considered very good.

"Diagnostic accuracy for exclusion of critical coronary artery stenosis over 50% indicates that 64-slice CT may be beneficial in triage of patients with atypical angina," Savino said.

Results by segment were weaker. In segments with stenosis of over 50%, accuracy was 96%, sensitivity 87%, and PPV 70%. In segments with stenosis of over 70%, accuracy was 98%, sensitivity 86%, and PPV 49%.

The study shows that 64-slice scanning by segment is better in comparison with earlier-generation CT systems, but there is still room for improvement.

"Sixty-four-slice CT coronary angiography provides sufficient sensitivity and negative predictive value to rule out significant coronary stenosis in patients with suspected CAD. But on a per-segment basis, sensitivity and diagnostic accuracy are still impaired, due to limited spatial resolution," Savino said.

CTA using the 64-slice system did produce a 100% negative predictive value on a per-patient basis, however, said Dr. U. Joseph Schoepf, an associate professor of radiology and medicine at MUSC. As a result, there is no negative effect on patient management.

"Since we are treating patients and not segments or vessels, there is very little to no risk that a patient with significant stenosis will not be detected by 64-slice coronary angiography," he said. "Thus, coronary CTA is shaping up as a highly sensitive triage tool for noninvasive rule-out of significant disease in patients with equivocal findings who otherwise would have to undergo invasive cardiac catheterization."