A comparison of the image quality of full-time myocardial perfusion SPECT vs wide beam reconstruction half-time and half-dose SPECT

Abstract

Objectives

Wide Beam Reconstruction (WBR) (UltraSPECT, Ltd) uses resolution recovery and noise modeling to cope with decreased SPECT count statistics. Because WBR processing reconstructs half the usual SPECT count statistics, we postulate that image quality equivalent to a full-time acquisition can be achieved in either half the time or with half the radiopharmaceutical activity.

Methods

In 156 consecutive patients (pts) rest and 8-frame gated post-stress myocardial perfusion SPECT was performed following 333-444 and 1184-1480 MBq (9-12 and 32-40 mCi) Tc-99m sestamibi injections, respectively, with full-time (rest = 14 min; stress = 12.3 min) acquisitions processed with OSEM and also separate “half-time” acquisitions processed with WBR. A subsequent group of 160 consecutive pts matched in gender, weight, and chest circumference received “half-dose” rest and stress injections 214.6 ± 22.2 and 647.5 ± 92.5 MBq (5.8 ± 0.6 and 17.5 ± 2.5 mCi) with full-time SPECT acquisitions. Image quality (1 = poor to 5 = excellent) was judged by myocardial count density and uniformity, endocardial edge definition, perfusion defect delineation, right ventricular visualization, and background noise.

Results

Mean image quality for rest, stress, and post-stress gated images were 3.6 ± 0.7, 3.8 ± 0.7, and 3.9 ± 1.0, respectively, for “full-time OSEM; 3.7 ± 0.8, 4.0 ± 0.7, and 4.8 ± 0.4 for “half-time” WBR; and 4.3 ± 0.8, 4.6 ± 0.6, and 4.7 ± 0.6 for “half-dose” WBR. “Half-time” and “half-dose” WBR image quality were both superior to standard full-time OSEM (P’s < .001). There was no significant difference between the summed stress and rest scores for “full-time” OSEM vs “half-time” WBR in 82 patients with perfusion defects.

Conclusions

Both “half-time” and “half-dose” WBR provide myocardial perfusion SPECT quality superior to full-time OSEM, with an associated decrease in scan acquisition time and patient radiation exposure, respectively.

Introduction

Myocardial perfusion SPECT has been rightfully criticized both for the length of the imaging procedure and for the radiation dose a patient receives. The advantages of shortening image acquisition times include improved patient comfort and tolerance, a decreased opportunity for patient motion, and enhanced laboratory throughput. New camera/detector systems have been developed, incorporating semiconductor detectors and focused collimation, considerably increasing cardiac SPECT counting statistics and thereby allow for greatly reduced image acquisition times.¹ Another solution has been the introduction of software methods incorporating depth-dependent resolution recovery and noise-modeling that provide good image quality despite reduced counting statistics, thereby allowing for shorter acquisition times.^2,3 Our group has previously reported gated image quality equivalent to that achieved with a full-time SPECT acquisition using both “half time” Wide Beam Reconstruction® (UltraSPECT Ltd, Haifa, Israel) (WBR) or “half-time” Evolution for Cardiac® (General Electric Medical Systems, Waukesha, WI).⁴ Subsequently, we reported gated and summed tomogram image quality equivalent to that achieved with full-time SPECT for “quarter-time” WBR, incorporating a modified algorithm with enhanced noise modeling.⁵

However, in recent years the volume of myocardial perfusion SPECT scans performed in the United States has plateaued or decreased slightly, most likely due to a downturn in the economy and insurance pre-certification requirements. Therefore, the need for increased throughput has not been realized in many laboratories. In contrast, the medical community, the public, and the media have demonstrated an increased awareness and have expressed caution regarding patient radiation exposure from diagnostic medical procedures. Diagnosticians have been urged to decrease and limit patient radiation exposure whenever possible. Moreover, because of the recent downtime of several nuclear reactors that produce Mo-99, there was a shortage of Tc-99m, also prompting laboratories to decrease radiopharmaceutical activity.

Since the new software methods referenced above generate good quality images with lower counting statistics, alternatively it should be possible to decrease the myocardial perfusion radiopharmaceutical activity while maintaining the standard SPECT acquisition time to generate equally acceptable gated and summed tomographic images. We therefore postulate that with either a reduced-time or reduced-activity acquisition yielding the same counting statistics, reconstructed cardiac SPECT images should be of equivalent quality.