Monday, November 12, 2007

Some Thoughs about SPECT/CT Scanners

Many of our sites in the States have acquired or are in the process of obtaining SPECT/CT scanners, a hybrid device philosophically similar to the now ubiquitous PET/CT’s. The reasoning behind both combinations is similar. In each case, it is desirable to form an accurate attenuation map, which CT will do far better than some external gamma-source. But more important to the interpreting radiologist, there is an exact match created between the SPECT or the PET, and the CT, facilitating diagnosis.
There are several options for SPECT/CT. General Electric makes the Hawkeye Infinia 4 with a limited, low power CT, used exclusively for attenuation correction. The Siemens Symbia pairs a full CT with a dual-head gamma camera. The Philips Precedence attaches a rather large articulated dual head SPECT camera to an equally large CT.
Our sites have a smattering of all three units. Images from the Precedence and the Symbia are quite good, and in particular, localizing lesions is made much easier with the synchronized SPECT and CT exams. Of course, reading these studies remotely presents other problems, but they are surmountable. Imaging from ProstaScint, OctreoScan, and even the occasional routine SPECT bone and cardiac study yields quite remarkable results.
The Hawkeye has limited CT capability. It can acquire 4 5mm slices simultaneously, but certainly the gantry must not rotate very quickly as it is an open configuration. The mAs and maximum kVp are simply not up to the level of the other two systems. The CT data is adequate for attenuation correction, but the images produced are far from diagnostic. Thus, one must get an additional diagnostic CT, eliminating the dose savings claimed by the maker of the Hawkeye. Their own literature even recommends fusion with a diagnostic CT. But such fusion is not always easy, as a photo from GE’s white paper on Symbia suggests:

Notice that the renal activity does not match the anatomic location of the kidneys. We are to assume that the hot focus does indeed correspond to a periaortic node, but how can we be certain? Granted, one can with Xeleris software perhaps do a better job of fusing the images, but should that be necessary? I think not. There are many problems inherant in fusing two separate examinations, and these are solved by the sequential acquisition found with those scanners that include diagnostic CT capability. Our sites that use the Hawkeye do sadly report that despite rather aggressive marketing, and assurances that the CT situation would make no difference, the Hawkeye’s CT images are indeed suboptimal, and a diagnostic CT is truly required.

I must conclude that although the Hawkeye is the least expensive of the three major contenders in this sector, its limitations outweigh the savings. The choice between the Siemens and the Philips probably then comes down to personal preference as well as the level of service each company can provide in your area.

No comments: