Reduced isotope dose and imaging time with a high-efficiency CZT SPECT camera
- First Online:
- Cite this article as:
- Duvall, W.L., Croft, L.B., Ginsberg, E.S. et al. J. Nucl. Cardiol. (2011) 18: 847. doi:10.1007/s12350-011-9379-7
- 596 Views
In light of recent focus on diagnostic imaging, cardiac SPECT imaging needs to become a shorter test with lower radiation exposure to patients. Recently introduced Cadmium Zinc Telluride (CZT) cameras have the potential to achieve both goals.
During a 2-month period patients presenting for a Tc-99m sestamibi SPECT MPI study were imaged using a CZT camera using a low-dose rest-stress protocol (5 mCi rest and 15 mCi stress doses). Patients ≥250 lbs or a BMI ≥35 kg/m2 were excluded. Rest images were processed at 5- and 8-minute acquisition times and stress images at 3- and 5-minute acquisition times. A subset of patients had stress imaging performed using both conventional and CZT SPECT cameras. Image acquisition times and SPECT camera images were compared based on total counts, count rate, image quality, and summed rest and stress scores. Twelve month clinical follow-up was also obtained.
131 patients underwent the study protocol (age 64.9 ± 9.8 years, 54.2% male). There was no significant difference in image quality and mean summed scores between 5- and 8-minute rest images and between 3- and 5-minute stress images. When compared to a conventional SPECT camera in 27 patients, total rest and stress perfusion deficits and calculated LVEF were similar (r = 0.94 and 0.96, respectively). At 12 months there was a benign prognosis in patients with normal perfusion. The effective dose was 5.8 mSv for this protocol which is 49.2% less than conventional Tc-99m studies and 75.7% less than conventional Tl-201/Tc-99m dual isotope studies.
New SPECT camera technology with low isotope dose significantly reduces ionizing radiation exposure and imaging times compared to traditional protocols while maintaining image quality and diagnostic accuracy.
KeywordsMyocardial perfusion imaging: SPECTradiopharmaceuticalsdiagnostic and prognostic applicationimage quality
Medical radiation exposure remains a topic of discussion as medical imaging has continued its rapid growth.1,2 The National Council on Radiation Protection and Measurements reported that the per-capita effective radiation dose of the United States population increased 72% from the early 1980s to 2006 which was felt to be the result of a 5.7 times increase in medical imaging.3 In 2006, the cumulative dose from non-therapeutic radiation included a contribution of 25% from all radiographic fluoroscopic procedures, 49% from computed tomography (CT), and 26% from nuclear medicine.4 As more attention has turned to the lack of monitoring and paucity of data on longitudinal radiation exposure in patients who frequently have multiple procedures performed over time, the risks and benefits of these procedures are being debated.4
Although extensively utilized in the management of patients with known or suspected coronary artery disease (CAD), myocardial perfusion imaging (MPI) suffers from known limitations, most notably long image acquisition times and radiation exposure. Recent studies of medical imaging have shown that MPI studies have one of the largest contributions to cumulative radiation dose.2 A recent American Society of Nuclear Cardiology (ASNC) statement and Food and Drug Administration white paper on reducing radiation exposure in MPI emphasize not only ensuring appropriate testing as a way of limiting exposure, but also adjusting stress protocols, limiting radiotracer dose, and using new technologies.5,6 The use of stress-only imaging which avoids the rest isotope dose has been shown to have an equivalent prognosis to conventional rest-stress studies particularly in patients without a known history of CAD7,8 and can be performed with lower stress doses.9 Recently introduced camera systems with optimized acquisition geometry, collimator design, and reconstruction software have the potential to allow reduced isotope doses while maintaining image quality with significantly shorter image acquisition times.10 The Discovery NM 530c (GE Healthcare, Haifa, Israel) high-efficiency cardiac camera is based on the multi-pinhole design and an array of Cadmium Zinc Telluride (CZT) pixilated detectors. The use of CZT detectors improves the energy and spatial resolution while the use of simultaneously acquired views improves the overall sensitivity and gives complete and consistent angular data.10
By reducing study time and decreasing radiation exposure, new CZT camera technology may allow SPECT MPI to more effectively compete with other imaging modalities for the diagnosis of CAD and addresses valid concerns about ionizing radiation from medical imaging. Reducing the long-standing, traditional 10/30 mCi Tc-99m rest-stress doses to our knowledge has not previously been studied. The objective of this study was to evaluate the feasibility of a low-dose rest-stress protocol with CZT camera technology by examining the image quality, quantitative perfusion assessment, and clinical outcomes of patients undergoing a new imaging protocol.
We prospectively evaluated all patients who presented to the Mount Sinai Non-Invasive Cardiology Laboratory over a 7-week period (October 2009 to December 2009) for their first clinically indicated Tc-99m sestamibi gated SPECT MPI stress test. Patients over the age of 40, <250 lbs, with a BMI <35, and who were not pregnant underwent rest and stress imaging on a CZT camera (Discovery NM 530c) using a low-dose protocol. A rest-stress imaging sequence was employed using Tc-99m sestamibi with a rest dose of 5 mCi and a stress dose of 15 mCi. When feasible, the first patient of the morning and afternoon sessions also had stress imaging performed on a traditional Anger SPECT camera (Vertex Plus, Philips/ADAC Laboratories) after their CZT camera image acquisition.
Imaging and Stress Protocol
A rest-stress imaging sequence was employed using Tc-99m sestamibi. The Tc-99m rest dose was 5 mCi and patients were imaged on the CZT camera in List Mode for 8 minutes allowing image processing at 5 and 8 minutes. The Tc-99m stress dose was 15 mCi and patients were imaged on the CZT camera in List Mode for 5 minutes allowing image processing at 3 and 5 minutes. Image acquisition began 30-60 minutes after tracer injection. All stress images were gated. Left ventricular ejection fraction (EF) was determined using commercial software (QGS, Cedars-Sinai, Los Angeles, CA, USA).
The Discovery NM 530c is equipped with a multiple pinhole collimator and 19 stationary CZT detectors simultaneously imaging 19 cardiac views. System design enables imaging of a three-dimensional volume imaged simultaneously by all detectors. Patients were imaged in the supine position with arms placed over the head. Automated heart positioning in the quality field-of-view was assisted using real-time persistence imaging. Although the detector could be rotated by the gantry if required for positioning, once acquisition was started there was no detector or collimator motion. Penalized maximum likelihood iterative reconstruction adapted to the Discovery NM 530c three-dimensional geometry was used to create transaxial slices of the left ventricle. No correction for scatter or attenuation was performed.11
A standard imaging protocol as endorsed by ASNC was used for stress images on the conventional SPECT camera.12 Gated SPECT stress imaging was performed using a dual head camera, step and shoot acquisition with 32 stops (16 frames per head), a 180° arc from right anterior oblique to left anterior oblique, a 64 × 64 × 16 matrix, and a VXGP collimator. Stress image acquisition began immediately following image acquisition on the CZT camera and images were acquired for a total of approximately 15 minutes (50 seconds per stop for 16 stops). All stress images were gated. Left ventricular EF was determined using the same commercial software (QGS, Cedars-Sinai, Los Angeles, CA, USA).
Organ doses were estimated using the most recent International Commission on Radiological Protection (ICRP) dose coefficients for each radiopharmaceutical, as reported in ICRP Publication 8015 for Tc-99m sestamibi, and in ICRP Publication 10616 for Tl-201. Effective doses were determined from organ doses using tissue weighting factors specified in ICRP Publication 103.17
Patient demographics, stress test variables, and tracer doses were prospectively collected at the time of stress testing in the Nuclear Cardiology Database. Pretest risk of coronary disease was based on ACC/AHA scoring system which utilizes age, gender, and presenting symptom.18
Image quality was graded on a subjective four-point scale (1 = poor, 2 = adequate, 3 = good, and 4 = excellent) by two readers and an average of score was calculated. Quantitative perfusion scoring of the rest and stress images was done by consensus and performed using a 17-segment model and a five-point scale (0 = normal, 1 = mild, 2 = moderate, 3 = severe, and 4 = absent).19 The percent total perfusion deficit was calculated by dividing either the summed rest or stress score by 68 and multiplying by 100.20 All reads were performed by two out of three Board Certified nuclear cardiologists (WLD, LBC, MJH) who were blinded to the imaging acquisition time. Inter-observer variability was calculated using the entire cohort as all patients were evaluated by two readers. A subset of patients with a mixture of normal (20%), probably normal (20%), mildly abnormal (16%), and abnormal (44%) perfusion scans was randomly chosen and read a second time 2 weeks later for intra-observer variability. Stress imaging time and total counts in the field of view were recorded. If the summed stress score was ≤3, the study was considered normal.
Patients were contacted by phone at 3 months, 6 months, and 1 year after their stress MPI study to determine the occurrence of any clinical events including hospitalization, repeat diagnostic testing, revascularization (PCI or CABG), non-fatal myocardial infarction (MI), or death. If patients were unable to be contacted by phone, office and hospital records were searched. Clinical outcomes were compared based on MPI results (normal or abnormal).
Continuous variables are presented as mean ± SD. Comparisons among continuous variables were done using unpaired Student’s t tests for demographic variables and paired Student’s t tests for comparisons between imaging times. Chi-squared tests were used to compare categorical variables. Correlation between cameras and observers was evaluated using linear regression, Pearson’s and Spearman’s rank method, as well as by intraclass correlation. A two-tailed P value <.05 was considered significant. Statistical analysis was performed using GraphPad Instat 3.1, Prism 5.0, and SPSS 19.
Entire cohort N = 131
Conventional SPECT cohort N = 27
64.9 ± 9.8
63.7 ± 10.1
26.2 ± 4.2
25.9 ± 4.3
Cardiac risk factors
Family h/o CAD
Shortness of breath
Stress MPI characteristics
Entire cohort N = 131
Conventional SPECT cohort N = 27
9.0 ± 2.7
8.8 ± 2.8
% Max HR
88.8% ± 8.6%
91.3% ± 10.6%
Ejection fraction (%)
63.8% ± 12.4%
62.6% ± 14.6%
Image Quality and Segmental Scoring
Image quality, segmental scoring results, and detected counts in the field of view using the CZT camera
Rest 5 minutes
Rest 8 minutes
Stress 3 minutes
Stress 5 minutes
3.16 ± 0.57
3.16 ± 0.61
3.62 ± 0.54
3.61 ± 0.54
Mean rest TPD (%)
8.3% ± 7.1%
8.2% ± 7.2%
Mean stress TPD (%)
6.0% ± 7.4%
5.9% ± 7.1%
1.047 ± 0.291
1.660 ± 0.430
1.823 ± 0.450
2.935 ± 0.615
Count rate (counts/second)
3488 ± 971
3459 ± 896
10128 ± 2497
9782 ± 2050
Intraclass correlation coefficients for inter- and intra-observer variability between short and long image acquisition times using CZT SPECT
Rest 5 minutes
Rest 8 minutes
Stress 3 minutes
Stress 5 minutes
Comparison to Conventional SPECT
MPI with high-efficiency CZT SPECT MPI compared to conventional SPECT MPI
High speed CZT SPECT
Stress 3 minutes
Stress 5 minutes
Stress 15 minutes
1.764 ± 0.403
2.851 ± 0.594
8.193 ± 1.377
Count rate (counts/second)
9798 ± 2237
9502 ± 1979
5284 ± 881
3.69 ± 0.46
3.61 ± 0.54
3.35 ± 0.6
Mean stress TDP (%)
7.8% ± 11.9%
7.4% ± 10.7%
6.8% ± 10.5%
Acquired Counts and Radiation Exposure
12-Month clinical follow-up results
Normal MPI N = 87
Abnormal MPI N = 44
Lost to follow-up
This study demonstrates that with new high-efficiency CZT SPECT camera technology a low Tc-99m dose protocol (5 mCi for rest and 15 mCi for stress) is feasible and effective radiation doses to patients can be reduced by 50%-75% compared to conventional doses. Decreased imaging time, from the traditional 15-20 minutes to only 5 minutes for rest imaging and 3 minutes for stress imaging, can also be achieved and is welcomed by patients, improves laboratory patient flow and productivity, and minimizes patient motion during imaging. When compared to the conventional SPECT camera, diagnostic accuracy and short-term prognosis are preserved. Thus, with new CZT camera technology, the goal of shorter tests and lower radiation dose appears to be feasible.
There have been several published studies on high-speed myocardial perfusion imaging.11,20-25 The first camera system studied was the D-SPECT (Spectrum Dynamics, Caesarea, Israel) which uses CZT solid-state detector columns with wide-angle tungsten collimators. An initial study found high image quality with an equivalent level of diagnostic confidence with traditional tracer dose when compared with conventional SPECT,22 while a second study explored a rapid (<20 minutes) dual isotope (Tl-201 stress/Tc-99m rest) protocol.23 The most recent studies with the D-SPECT camera, again compared high-speed SPECT to conventional SPECT using standard isotope doses in a multicenter setting20 and explored simultaneous dual-radionuclide imaging.25 Initial studies using the Discovery NM 530c camera with conventional isotope doses found comparable diagnostic performance compared to conventional SPECT with significantly shorter acquisition times.11,21,24
As with previously published studies, we found similarly high image quality and excellent diagnostic performance with CZT SPECT.11,20,22,23 The protocol in this study allowed rapid imaging with better image quality compared to a conventional SPECT camera. The diagnostic performance as measured by total perfusion deficits showed excellent correlation between long and short image acquisition times and when compared to conventional SPECT. This is also the first study to incorporate clinical follow-up of patients who underwent CZT SPECT imaging. Among patients with normal myocardial perfusion, at 12 months there were no reported cardiac deaths and only two coronary revascularization which is similar to prior studies which have demonstrated cardiac event rates of <1% after a normal conventional SPECT MPI at 1 year.26,27
This is the first study examining low isotope doses for rest-stress Tc-99m CZT SPECT imaging. The high-efficiency CZT SPECT detector technology with its 5-8 fold increased sensitivity22,28 allows for both the rapid imaging and the reduction in isotope dose. While our previous work with high-efficiency SPECT cameras emphasized reduction in radiation exposure, it did so by lowering isotope dose in the setting of an abbreviated stress-only protocol.9 Another abbreviated stress-only protocol utilized conventional SPECT with half-time iterative reconstruction imaging for 15 minutes with half of the recommended low-dose isotope dose.29 The previous dual isotope imaging protocol by Berman et al23 employed a low 2 mCi Tl-201 stress dose and an 8 mCi Tc-99m rest dose, but focused more on the rapidity of the protocol than on maximizing reduction in effective dose. In fact, the effective dose of 11.9 mSv with the rapid dual isotope protocol is similar to that of a standard rest-stress Tc-99m protocol. Thus, the earlier CZT SPECT studies did not concentrate on radiation exposure reduction as was emphasized here, instead employing conventional Tc-99m doses or a dual isotope protocol.
The current protocol, to our knowledge, represents the lowest effective dose to patients for a Tc-99m rest-stress protocol studied to date. It is also feasible that the doses used in this study could be further reduced in select patients with appropriate body habitus. While reduction in radiation exposure may not be as relevant to the elderly or those with reduced life expectancy due to comorbid disease, the concept of ALARA (“As Low As Reasonably Achievable”) is especially pertinent in this day of multiple repeat studies which begin at an increasingly younger age. Because of the volume of cardiac diagnostic procedures involving ionizing radiation, most cardiac patients are exposed to an increasing cumulative dose and a reduction in effective dose of 49% compared to traditional rest-stress Tc-99m studies and 76% compared to traditional Tl-201/Tc-99m dual isotope studies is welcome. A low-dose isotope protocol would also make MPI more suitable for non-invasive hybrid cardiac imaging such as CT angiography-MPI because of the lower total effective dose.30 The ability to reduce the effective dose of a full rest-stress study to 5 mSv represents a major advancement in myocardial perfusion imaging’s ability to remain viable and competitive. Low-dose isotope protocols coupled with stress-first imaging protocols could further reduce radiation exposure to patients.7-9,29
The study is limited by the sample size and the single site clinical experience making it a pilot or feasibility study. This study did not compare low-dose CZT SPECT rest images to low-dose conventional SPECT rest images. However, the low tracer dose used in the rest images would likely require prohibitively long imaging time on a conventional SPECT camera. The more standard dose of 15 mCi for stress images allowed for a valid comparison between cameras with the stress images. Not all patients were imaged with the conventional SPECT camera for comparison and not all patients underwent coronary angiography as a gold standard. Patients with a body weight >250 lbs or BMI ≥35 kg/m2 were excluded from the analysis. Clinical follow-up was limited by the small sample size, short duration of follow-up, and lack of clinical events. The counts and count rate analysis utilized counts in the field of view not the more relevant counts in the myocardium. Because of the larger field of view with the conventional SPECT camera, the total counts and count rates are likely over estimated for that camera due to greater contribution from GI and extra-cardiac sources.
New CZT SPECT camera technology with low isotope dose, 5 mCi Tc-99m rest dose and 15 mCi Tc-99m stress dose, significantly reduces ionizing radiation exposure compared to traditional protocols while maintaining image quality and diagnostic accuracy. No difference in image quality was found using shorter 5 minutes (instead of 8 minutes) for rest imaging and 3 minutes (instead of 5 minutes) for stress imaging. Compared to conventional SPECT, image quality was better with CZT SPECT, and normal versus abnormal perfusion with the CZT SPECT camera effectively differentiated clinical outcomes. This study demonstrates that CZT SPECT can overcome the limitations of conventional SPECT imaging.