Thallium-201 and 99mTc-sestamibi images of the heart contain a significant amount of scattered events which degrade image quality. Newer generation gamma cameras exhibit enhanced energy resolution and hardware/software to perform scatter correction. The principal aim of this study was to evaluate the effects of these advances in instrumentation on the quantitation of defect size from tomographic images of the heart obtained from a cardiac phantom.
Methods: Tomographic images of a cardiac phantom containing no defect and defects of 5%-70% of total myocardial mass were acquired both with and without scatter correction for 201Tl and 99mTc studies. Data were acquired on a newer generation gamma camera with an energy resolution of 8.7% at 140 keV. From conventional short-axis slices of the heart, circumferential count profiles were generated from five representative slices. Defect size was computed from the fraction of radians that fell below a fixed threshold value in each of the five count profiles. The nadir value (min/max) of the count profiles in each study was used as an index of image contrast.
Results: For both 201Tl and 99mTc, threshold values between 55%-60% gave the best correlation (r > 0.99), with the lowest average absolute error in estimating defect size (< 2.1%). Scatter correction reduced the average absolute error to 0.8% for 99mTc and 1.4% for 201Tl, significantly reduced the nadir values for both isotopes (p < 0.0001 for both 201Tl and 99mTc and led to a marked improvement in image quality for both tracers.
Conclusion: Scatter correction reduces the error associated with measurement of infarct size, increases image contrast and improves image quality for both 201Tl and 99mTc, as assessed in a phantom model.