Children of a lesser god: From radiocardiogram to CZT GSPECT

Excerpt

We had the fortune to work in the Institute founded and directed for more than 40 years by Luigi Donato, a pioneer of nuclear medicine, among the most influential cardiologist of the past century. He loves to remember those days in the 1950s, when he, a young research fellow, drove every month from Pisa to Ciampino airport in Rome to wait the arrival of Na24. Then, he was used to put the box with the radioisotope into the mythical Italian car “Fiat Topolino” (i.e., “baby tiny mouse”), coming back to Pisa and projecting his research work. Professor Donato is one of the fathers of radiocardiography, a technique based on the radioisotopic application of the indicator-dilution principle, representing the first successful attempt to noninvasively obtain information on cardiopulmonary circulation1 (Figure 1). A few years later, Folse and Braunwald described a radioisotope indicator-dilution technique to assess the ‘fraction of left ventricular volume ejected per beat and ventricular end-diastolic and residual volumes’.2 It is still difficult to fully understand how these studies have influenced cardiology. Successively, cardiac imaging has evolved following technological developments. In nuclear medicine, the advent of Anger camera and the use of ^99mTc-labeled compounds led to implement cardiac blood-pool imaging, a gold standard technique for the assessment of biventricular function, till the validation of computed tomography and cardiac magnetic resonance (cMRI).3 Another revolution started in the 1970s with the diffusion of echocardiography: the investigation of cardiac pump function became relatively simple, immediate and tightly linked to the cardiologists’ world, with the significant limitation of a rather subjective, operator-dependent assessment. Perhaps, the success of cardiac ultrasound methodology has then forced nuclear cardiologists toward the nobody’s land of myocardial perfusion, with the scope of simultaneous assessment of myocardial perfusion/function. The dream has become reality at the end of the 1990s, with the work of Germano, Bermann and other researchers: the development of gated-SPECT, permitting the noninvasive evaluation of myocardial perfusion, viability and function in the same examination, regenerated nuclear cardiology.4 Since its implementation, gated-SPECT has promised obvious advantages over perfusion evaluation only. The simple visualization of wall motion and thickening has greatly improved the identification of attenuation artefacts, with a significant increase in specificity (and overall diagnostic accuracy) of myocardial perfusion SPECT.5 The possibility to investigate regional wall motion abnormalities furnished an incremental value in the identification of coronary artery disease and for the evaluation of its extent and severity.6 Measurements of left ventricular ejection fraction (LVEF) and volumes improved prognostic stratification.7 Last but not least, the measures derived were obtained by means of semiautomatic operator-independent algorithms, resulting in a high intra- and inter-observer reproducibility. However, validation studies performed against the gold standard cMRI showed significant differences in both LVEF and volume measures, with a tendency to underestimate EF values and end-diastolic volumes (particularly in patients with advanced heart failure) and to overestimate end-systolic volumes (particularly in small -female- hearts).8,9 These findings recognize a complex explanation, including count statistics, spatial and temporal resolution, differing between gated-SPECT and cMRI. Moreover, cMRI allows the inclusion of the outflow tract, which is not a part of LV volume acquisition with gated-SPECT, because of the absence of counts in this specific area. Recently, further technological developments have introduced digital imaging in nuclear cardiology. Semiconductor cadmium zinc telluride detectors ensure a striking improvement in spatial and energy resolution, associated with a dramatic increase of count statistics, a very important issue when using a technique like gated-SPECT for evaluating LV function, strongly dependent on counts detected in each phase of the cardiac cycle. Preliminary validations of CZT cameras vs cMRI have provided encouraging results, using either a standard dose 10 or a low-dose acquisition protocol.11 In the present number of the Journal, Plateau and colleagues 12 compared the relative performance of the available software packages for nuclear cardiology LV function assessment versus the gold standard cMRI. This is not a trivial problem when making quantitative measurements of a physiological parameter. Semiautomatic software packages guarantee reproducibility but, being based on different mathematical and geometric approaches, can result in increased variability of the obtained measures. Authors concluded that the 4 investigated software packages provided satisfactory results, when compared to cMRI, but with wide limits of agreement.

…