The right stuff

Comparison of radionuclide ventriculography using SPECT and planar techniques in different cardiac conditions

PURPOSE

Accurate assessment of ventricular function is required to optimize therapeutic management of cardiac diseases. The aim of this study was to correlate planar equilibrium multigated acquisition (MUGA) with tomographic ventriculography (SPECT) in patients with diverse volumes and wall motion abnormalities.

METHODS

Eighty-three studies in 80 patients (56+/-14 years; 56% women) were classified according to ventricular dilation, wall motion abnormalities and systolic dysfunction. Left and right ventricular ejection fraction (LVEF and RVEF) and end-diastolic and end-systolic left ventricular volumes (EDV and ESV) were obtained using a commercial QBS program for SPECT. On planar acquisition, LVEF and RVEF were obtained using standard techniques and volumes were determined using the count-based method, without blood sampling.

RESULTS

A. Total group: With the planar method, LVEF was 44+/-17%, RVEF 42+/-13%, left EDV 147+/-97 ml (range 31-487 ml) and left ESV 93+/-85 ml (range 15-423 ml); with SPECT the corresponding values were 40+/-20%, 49+/-16%,131+/-95 ml and 91+/-89 ml, respectively (p=NS for all but RVEF). Linear correlation was 0.845 for LVEF, 0.688 for RVEF, 0.927 for left EDV and 0.94 for left ESV, with good intra-class correlation. B. Subgroups: Global and intra-class correlations between planar imaging and SPECT were high for volumes, RVEF and LVEF in all subgroups, except in patients with normal wall motion and function, who showed smaller volumes with SPECT. The group with diffuse wall motion abnormalities had a lower EDV on SPECT. In the abnormal left ventricle, RVEF was higher with SPECT.

CONCLUSION

Good correlation and agreement exist between SPECT and planar MUGA with respect to LVEF and left ventricular volumes. SPECT is useful in patients with functional abnormalities, but less reliable in those with normal small cavities. A combined technique is still necessary, and RVEF should be interpreted cautiously.

Quantitative single-photon emission computed tomography imaging

Over the past decade, quantitation of cardiac single-photon emission computed tomography (SPECT) data, once limited to perfusion assessment, has been extended to global and regional function assessment for both the left and the right ventricle, as well as to measurement of additional cardiac parameters of diagnostic and prognostic interest. A number of commercially available quantitative algorithms exist, based on different mathematic operators and with varying degrees of automation, that are capable of providing accurate and reproducible results. This article describes the many quantitative cardiac SPECT measurements available today, defining them in terms of validation, practical use, and limitations.

Evaluation of right ventricular function by NuSMUGA software: gated blood-pool SPECT vs. first-pass radionuclide angiography

BACKGROUND

In comparison with planar imaging gated blood-pool single photon emission computed tomography (GBPS) has the advantage of separating left and right ventricle. The purpose of this investigation was to evaluate the right ventricular ejection fraction (RVEF) calculations by GBPS software ('NuSMUGA', Northwestern University, Chicago, IL) in comparison to first-pass radionuclide angiography (FPRNA).

METHODS

In 22 consecutive patients FPRNA and GBPS acquisition was performed. GBPS RVEF calculations were manually and automatically performed, using all gated short-axis-slices of the right ventricle.

RESULTS

Automatic RVEF assessed by GBPS did not correlate with conventional FPRNA (r = 0.40, p = 0.065). Mean FPRNA RVEF was 55 +/- 10% and mean GBPS automatic RVEF was 32 +/- 8%. Also manual GBPS RVEF did not correlate with conventional FPRNA (r = 0.41, p = 0.055). Mean RVEF measurement by manual GBPS was 33 +/- 8%. Manual GBPS RVEF values correlated well with automatically determined GBPS RVEF values (r = 0.96, p < 0.0001).

CONCLUSION

Automatic and manual values RVEF values assessed by GBPS did not correlate with conventional FPRNA values. FPRNA and GBPS calculations cannot be considered to be equivalent. Therefore the NuSMUGA program cannot be used to calculate RVEF.

Automatic generation of noise-free time-activity curve with gated blood-pool emission tomography using deformation of a reference curve

This paper describes a new method for assessing clinical parameters from a noisy regional time-activity curve (TAC) in tomographic gated blood-pool ventriculography. This method is based on a priori knowledge on the shape of a TAC, and shape approximation. The rejection method was used to generate different random Poisson deviates, covering standard count levels, of six representative TACs in order to test and compare the proposed method with harmonic and multiharmonic reconstruction methods. These methods were compared by evaluating four clinical parameters: time of end systole, amplitude, peak ejection and filling rates. Overall, the accuracy of assessment of these parameters was found to be better with the method described in this paper than with standard multiharmonic fits.