Key points
- •
Artifactual perfusion defects may develop in the AC image if the CT image is not correctly registered with the SPECT image prior to reconstruction. Therefore, the CT image should be quality controlled prior to SPECT image interpretation.
- •
The radiation exposure from CT attenuation is small.
- •
AC eliminates the dependence of body habitus on the normal count distribution of myocardial perfusion images. As a result, the specificity and the normalcy rate of AC images are superior to non-AC images.
- •
Line sources for simultaneous transmission and emission imaging are the second most common alternative method for attenuation correction.
- •
Noncontrast CT images may be used to assess the presence and extent of coronary artery calcium. Because coronary calcium is pathognomonic of CAD, it could change the risk stratification of patients without known CAD who have a normal SPECT MPI.
- •
The absence of coronary artery calcium is associated with a very low (< 5%) probability of inducible myocardial ischemia on SPECT imaging. This information should be used in conjunction with the patient’s pretest probability of CAD to increase the diagnostic confidence of the physician in excluding obstructive CAD.
- •
The radiation exposure from coronary calcium scoring is small.
- •
SPECT MPI may underestimate the ischemic burden in patients with multivessel obstructive CAD because of the relative nature of perfusion distribution. In such cases, CT coronary angiography may help determine the true extent of myocardium at jeopardy.
- •
It may be difficult to determine the culprit artery in patients with multivessel CAD and inducible ischemia (such as in the anterolateral [LAD or LCX territory] or inferolateral [RCA or LCX territory]) walls. Software fusion of SPECT and CT angiography images may prove useful in isolating the culprit vessel.
- •
The radiation exposure from combined SPECT perfusion and CT angiography is of concern and might be a reason against its routine use.
Background
Hybrid SPECT/CT cameras are becoming more available in the nuclear cardiology community. These two imaging modalities, whether or not performed in the same setting, can pave the way to a more comprehensive assessment of CAD. While CT imaging allows detection and localization of anatomic CAD, SPECT imaging reflects the functional consequences of the disease process. Non-contrast CT images can be used for SPECT AC and coronary artery calcium scoring and contrast-enhanced images are used for CT coronary angiography. The clinical and economic effects of this technology have not yet been established. Optimal clinical use of hybrid imaging requires the appropriate selection of patients and an understanding of SPECT/CT quality control issues.
The attenuation image used for SPECT AC is obtained either with gadolinium line sources or with a CT scanner. While the gadolinium attenuation image is obtained simultaneously with the emission image, the CT attenuation image is acquired either before or after the emission image, increasing the likelihood of misregistration between the emission and transmission images. Artifactual perfusion defects may develop in the AC image if the CT image is not correctly registered with the SPECT image prior to reconstruction.
Case 19-1
SPECT/CT Registration Artifact: LV–Lung Mismatch ( Figure 19-1 )
A 60-year-old woman reported an episode of a heavy feeling in her chest lasting 5 to 10 minutes 2 weeks prior to an outpatient clinic visit. Chest pain was not associated with exertion and resolved spontaneously. Her only risk factor for CAD was hypercholesterolemia. Her physical examination was unremarkable. She exercised for 5 minutes on the treadmill, achieving 92% of maximum predicted heart rate. The ECG was unremarkable at baseline. Stress ECG showed no ischemic changes. The rest/stress (5/20 mCi) sestamibi SPECT images were obtained using a solid-state-detector camera equipped with multiple pinholes. A single CT attenuation image was obtained prior to rest SPECT imaging using a stand-alone 64-row CT scanner. The CT image was used for attenuation correction of the rest and stress SPECT images.
Comments
Emission-transmission misregistration of the stress images, but not the rest images, resulted in a large artifactual reversible perfusion defect in the AC SPECT images. Significant count losses in the myocardium are usually seen when the LV wall on the SPECT image overlays lung tissue on the CT image. The quality of the CT image can have a substantial impact on the quality of the AC SPECT image. For example, CT images with streaking artifacts and dark bands, which are commonly seen in patients with implantable metal devices (such as defibrillator leads), may result in artificially increased myocardial uptake in the corresponding AC SPECT image because of incorrect scaling of the Hounsfield units into the attenuation map. Also, hypoinflated or hyperinflated lungs on CT can be a source of artifacts in the AC SPECT images (see Figure 20-2 ). The CT and fused images should always be available at the time of SPECT image interpretation to ensure CT image quality and proper alignment of the emission and transmission images.
Case 19-2
SPECT/CT: Improving Image Quality With Attenuation Correction ( Figure 19-2 )
A 45-year-old man presented to the outpatient clinic with a history of daily atypical chest pain in the past month. He had no accompanying shortness of breath, nausea, or vomiting. His risk factors for CAD included a 10-pack-year history of smoking and a family history of CAD. The physical examination was unremarkable (215 lb, BMI 28 kg/m 2 ). He underwent vasodilator stress testing with 0.4 mg IV regadenoson. The baseline ECG showed normal sinus rhythm and first-degree AV block. Stress ECG showed no ischemic changes. The rest/stress (5/20 mCi) sestamibi SPECT images were obtained using a solid-state-detector camera equipped with multiple pinholes. A single CT attenuation image was obtained prior to rest SPECT imaging using a stand-alone 64-row CT scanner. The CT image was used for AC of the rest and stress SPECT images. Emission and transmission images were well-registered prior to AC SPECT image reconstruction.
Comments
Myocardial perfusion SPECT images are susceptible to soft tissue attenuation artifacts. These artifacts can be misinterpreted as perfusion defects and increase the false-positive rate of SPECT imaging. AC images have higher LV count homogeneity in normal patients than non-AC images because AC eliminates the dependence of body habitus on the normal count distribution. AC algorithms include not only correction for photon absorption, but also correction for scatter and for the degradation of resolution with depth. The correction for these phenomena improves image quality and increases the interpretive certainty of the physician.
Compensation for soft tissue attenuation and scatter resulted in a more homogeneous count distribution in the LV and is not consistent with hemodynamically significant CAD. The fixed count reduction in the inferior wall was attributed to diaphragmatic attenuation because, in addition to resolution of the defect on AC images, there was normal inferior wall motion on gated images and there were no Q-waves on the ECG.
There is increasing evidence that the specificity and the normalcy rate of AC images are superior to non-AC images. However, one should not expect that AC will yield perfect results every time. AC SPECT image quality may be compromised because of CT artifacts, truncation, misregistration, etc., which should all be quality controlled prior to AC SPECT interpretation. Today, the American Society of Nuclear Cardiology recommends AC as an adjunct to myocardial perfusion SPECT studies, whenever available.
Case 19-3
SPECT/CT: Improving Risk Stratification With Coronary Artery Calcium Scoring ( Figure 19-3 )
A 56-year-old man without known CAD presented to the outpatient clinic having suffered 2 episodes of atypical chest pain over the past week. The chest discomfort was not associated with exertion and resolved spontaneously. There were no associated symptoms. His risk factors for CAD included a 40-pack-year history of smoking and a strong family history of CAD (father had fatal MI at age 45). The physical examination was unremarkable. He was exercised using the standard Bruce protocol. The baseline ECG was normal. He exercised for 11 minutes, reached target heart rate, and stopped because of fatigue. Stress ECG showed no ischemic changes. The rest/stress (10/30 mCi) sestamibi SPECT images were obtained using a standard dual-detector SPECT camera covering a 180-degree imaging arc from the 45-degree RAO projection to the 45-degree LPO projection. A non-contrast breath-hold ECG-gated chest CT was obtained prior to rest SPECT imaging, using a stand-alone 64-row CT scanner.
