Combined SPECT/CT scanners are receiving a warm welcome in the nuclear cardiology community. Acquisition of CT data immediately before or after a myocardial perfusion SPECT exam is proving to be an effective means of minimizing attenuation artifacts. Multislice SPECT/CT units could also pave the way to comprehensive assessments of cardiac function.
SPECT images are inherently susceptible to attenuation artifacts. Differing scatter and absorption of photons between emission and detection can distort the final picture. Failure to correct for likely attenuation along the photon's path can cause spatial misrepresentation.
A key problem for cardiac SPECT is that soft-tissue attenuation artifacts may be confused with perfusion defects. Attenuation artifacts can also obscure genuine coronary artery disease. CT-derived anatomical maps are finding increasing favor as a tool for correcting likely photon attenuation and improving diagnostic accuracy.
"Right now, we are doing all of our cardiac studies with SPECT/CT for attenuation correction," said Prof. Ora Israel, director of nuclear medicine at the Rambam Medical Center in Haifa, Israel. "Hybrid SPECT/CT was first developed because we were looking for a better way to do attenuation correction for cardiology, and to this day x-ray attenuation correction is the best method."
Nuclear cardiologists are eager to see how the addition of multislice technology to SPECT/CT hybrids affects attenuation correction. At Johns Hopkins University School of Medicine, for instance, CT attenuation data are routinely used to improve the accuracy of myocardial perfusion images. Until earlier this year, these exams were performed on one of four gamma cameras with single-slice CT. Radiologists now have the option of using a 16-slice SPECT/CT system instead.
"We use the CT scan to make our SPECT scan more quantitatively accurate," said Dr. Richard Wahl, director of nuclear medicine/PET at Johns Hopkins. "It is used to tell how thick the patient's myocardium is and to eliminate false-positive results that might be generated by thinking that soft attenuation due to breast tissue or fat is a perfusion defect."
Dr. Michael O'Connor, a professor of radiologic physics at the Mayo Clinic in Rochester, MN, shares the view that CT data offer the best option for attenuation correction. Alternative methods that use external radiation sources to generate a transmission scan suffer from poor image quality, in part because of the low radiation flux of sources such as gadolinium-153, he said. Multislice SPECT/CT systems should acquire attenuation correction data far faster than the single-slice hybrids, which can take 13 seconds per slice.
"If you were doing a cardiac study on one of these older systems, it could take you five or 10 minutes. The newer devices take just a few milliseconds per slice, so to complete the CT component is on the order of seconds rather than minutes," O'Connor said.
The rapidity of the CT scan could have its own problems, however. CT data acquired in an instant will capture the heart frozen in one part of the cycle, but the heart will beat continuously during the SPECT part of the exam, which could take up to 20 minutes. Registering the averaged SPECT image with the fixed CT slice could in itself generate artifacts, he said.
The move to MSCT has not caused such problems at the University of Michigan Medical Center, however, where nuclear cardiologists have access to a SPECT/CT hybrid with integrated six-slice spiral CT. Collection of CT data for attenuation correction takes just five to 10 seconds on the new machine, and the diagnostic quality of the myocardial perfusion images has improved considerably, said Dr. James Corbett, director of cardiovascular nuclear medicine at the Michigan center.
"We have compared results from the new scanner with the SPECT systems we already have, which are equipped with CT for attenuation correction using the currently more common sealed source approach, and the news scanner clearly appears to work better," Corbett said. "The results are only preliminary, but we have been pleased with the attenuation correction in patients who had known disease."
The hybrid scanner's ability to correct for attenuation and scatter should also improve MIBG SPECT of patients with heart failure, he said. SPECT alone can be difficult to interpret because of spillover scatter from MIBG uptake in the liver and lungs.
Attenuation correction is not the only role that nuclear cardiologists are earmarking for multislice SPECT/CT. Several centers also plan to collect information on coronary calcium while patients are still on the table. The calcium scan will add diagnostic and prognostic information to the perfusion data. The approach should save time and provide near-perfect registration of coronary calcium with perfusion data, Corbett said.
"Studies have shown that coronary calcium scoring combined with SPECT perfusion imaging does provide additional value," he said. "And if one scan provides the attenuation correction as well, the value is further incremented."
Dr. Jack Ziffer, chief of radiology at Miami Baptist Hospital in Florida, is also interested in using diagnostic-quality CT to map calcium in patients undergoing myocardial perfusion SPECT. Miami Baptist is in the process of upgrading its nuclear medicine facilities to include two multislice SPECT/CT systems. The move to multislice CT offers the chance to trial new cardiovascular protocols.
"We've got the opportunity to do something that is fundamentally very different. That's the big excitement. This is why we thought we needed two of these machines," Ziffer said.
Nuclear SPECT has become the mainstay of diagnostic evaluation for patients presenting at Miami Baptist with suspected acute coronary syndromes. But the protocol is not entirely satisfactory, according to Ziffer. Under a proposed new protocol, this patient group would still undergo rest SPECT imaging, but all studies would be corrected for attenuation using multislice CT data. Those CT data would also be used for calcium scoring. Patients with a normal or equivocal resting SPECT scan and a calcium score of zero would be sent home without need for a stress test. Those with an abnormal or equivocally abnormal resting SPECT and prior clinical history that makes acute coronary syndrome unlikely could be immediately followed up with a CT coronary angiogram.