Chapter 25: Hybrid Imaging: SPECT–CT and PET–CT

There has been a significant growth in hybrid single-photon emission computed tomography and computed tomography (SPECT–CT) and positron emission tomography computed tomography (PET–CT) systems over the past decade, driven in large part by oncologic imaging. A fortuitous byproduct of this has been the development, application, and validation of myocardial perfusion imaging (MPI) using these hybrid systems. Cardiac hybrid SPECT–CT and PET–CT systems offer distinct advantages compared with traditional SPECT or PET MPI. The CT provides for excellent attenuation correction and improves the sensitivity and specificity of MPI, CT-derived coronary artery calcium (CAC) score adds substantial incremental diagnostic and prognostic information to MPI, and hybrid scanners offer the ability to combine a physiologic assessment of perfusion, function, or metabolism with an anatomic assessment of atherosclerosis and structural heart disease. By doing so, hybrid SPECT–CT and PET–CT imaging offers unprecedented opportunities for molecular cardiology research. The primary focus of this chapter is to discuss the clinical applications of hybrid radionuclide MPI with calcium scoring and coronary CTA.

The hardware of SPECT–CT and PET–CT scanners comprises a conventional SPECT scanner or a PET scanner coupled with a CT scanner of various configurations. While all SPECT–CT and PET–CT scanners offer CT-based attenuation correction, calcium scoring (≥4 slice MDCT) and coronary CTA (≥64 slice MDCT) may be performed only on certain hybrid SPECT–CT and PET–CT scanners. Sample hybrid PET–CT and SPECT–CT protocols are shown in Figure 25-1A and B, respectively. The CAC score and/or coronary CTA study can be performed sequentially right before or after the SPECT or PET scan or at a separate setting.

Attenuation correction using transmission scanning employing external radioactive sources or cardiac CT improves the count uniformity of the image and helps distinguish attenuation artifacts from real defects. It also offers the possibility of stress-only imaging, with potential savings of time, cost, and radiation dose as discussed in more detail in Chapter 9. Also, accurate attenuation correction enables precise measurements of absolute radiotracer concentration in the myocardium making feasible noninvasive quantitative estimation of myocardial blood flow in mL/g/min.

External radionuclide source transmission scans pose some challenges including: (1) degradation of the radionuclide source over time which adversely impacts image quality, (2) longer acquisition time when compared with the CT transmission scans, and (3) a small additional radiation dose (in addition to the emission scan). CT attenuation correction, on the other hand, is rapid (takes a few seconds) and of excellent quality. SPECT–CT and PET–CT utilize a low-dose x-ray transmission computed tomogram for attenuation correction (CTAC). But, due to inherent differences in image resolution between CT and SPECT or PET MPI, the CTAC and SPECT or PET myocardial perfusion images may be misregistered (not appropriately aligned) resulting in artifacts. Accurate registration is critical for improving the diagnostic yield of CT attenuation-corrected MPI (Fig. 25-2A and B).

SPECT–CT MPI has been validated in patients with and without underlying CAD.^1,2 Multiple studies of SPECT MPI with radionuclide attenuation correction compared with non–attenuation-corrected images demonstrated that test specificity and normalcy are improved while maintaining sensitivity to detect obstructive CAD.^3–6 Importantly, attenuation correction of MPI increases the normalcy rate, a term used to define the percentage of normal studies in a low-risk cohort. Figure 25-3 demonstrates the effect of attenuation correction using a hybrid SPECT–CT system. Recent publications have demonstrated the improved prognostic capability of attenuation-corrected SPECT MPI.^7,8 However, radionuclide attenuation correction with SPECT MPI is not easy to use and widespread clinical adaptation of this technique has been slow.

In contrast, PET MPI with attenuation correction has been widely used clinically (Fig. 25-4) and for research applications, but PET MPI without attenuation correction is significantly degraded by attenuation and is not interpreted clinically or for research purposes.⁹ Multiple studies demonstrated the excellent diagnostic¹⁰ and prognostic value¹¹ of PET MPI. Furthermore, noninvasive quantitative assessment of myocardial blood flow with PET has emerged as a powerful tool to diagnose microvascular dysfunction, follow progression or regression or CAD, and identify localized ischemia, transplant vasculopathy, and balanced ischemia.⁹ Absolute myocardial blood flow and coronary flow reserve assessed by PET accurately predicts future adverse cardiovascular outcomes; in particular, a normal coronary flow reserve offers excellent negative predictive value to exclude high risk CAD.^12–15

Advances in multidetector row CT scanners have tremendously improved the noninvasive diagnosis of CAD using CAC score, coronary CTA, combined imaging of myocardial perfusion and anatomy,¹⁶ as well as CT-derived estimates of fractional flow reserve (CT-FFR).¹⁷ For attenuation correction, low-resolution CT (nondiagnostic CT), 2-slice CT or ≥4-slice multidetector-row CT-based hybrid scanners can be used. For CAC scoring, at least 4-slice CT is required (≥6-slice recommended). For coronary CTA, at least a 16-slice scanner is required (≥64-slice multidetector-row CT recommended), with imaging capability for slice width of 0.4 to 0.6 mm and temporal resolution of 500 ms or less (≤350 ms is preferred).¹⁸

Coronary artery calcification is a specific marker of coronary atherosclerosis and a powerful indicator of increased cardiovascular risk (see Chapter 28).^19,20 CAC score in addition to MPI can be of great value both to the clinician and the patient and stimulate further discussions regarding patient's risk factor management.²¹ Likewise, extensive literature (as discussed in Chapter 28) supports the role of coronary CTA alone or in conjunction with MPI in the diagnosis and management of individuals with known or suspected CAD. Extensive coronary artery calcification, high or irregular heart rates, and high body mass index may limit the diagnostic accuracy of coronary CTA; ischemia evaluation may provide incremental diagnostic value in those instances. Furthermore, revascularization decisions guided by functional testing with SPECT MPI²² or invasive fractional flow reserve²³ may improve clinical outcomes.

Hybrid CAC Score and MPI

With hybrid SPECT/CT and PET/CT systems a prospectively gated calcium score scan can be acquired along with the MPI or the CTAC obtained during PET–CT or SPECT–CT MPI can be visually assessed for coronary artery calcification with good accuracy.^24,25 Several studies wherein subjects underwent both a CAC score study and MPI (at the same setting or at different settings) have demonstrated that subjects with normal MPI may have extensive underlying calcified atherosclerosis, and this finding may influence physicians to prescribe aspirin and lipid-lowering agents.²⁶ The frequency of ischemia in subjects with Agatston calcium score of >400 is high (>20%)^27–31; a myocardial perfusion study is considered appropriate among individuals with CAC score >400 or among individuals with high CHD risk and CAC score 100 to 400 independent of symptoms.³² As with CAC score, investigators³³ have evaluated the diagnostic and prognostic value of a zero CAC score in conjunction with stress SPECT MPI in patients presenting to the emergency room with chest pain. In that study, 0.8% of patients had an abnormal MPI (5/625 patients, four of whom had no CAD on subsequent invasive angiography), and 0.3 event rate (mildly elevated troponin, no cardiac death) over a mean follow-up of 7 months. These authors³³ concluded that most of the patients with chest pain in the ED have a calcium score of 0, which predicts both a normal stress SPECT result and an excellent short-term outcome. A recent meta-analysis by Bavishi et al.³⁴ confirmed that zero to low CAC scores were infrequently associated with ischemia, but there was a wide variance in the frequency of ischemia among patients with intermediate to high CAC scores (Tables 25-1 and 25-2).

Indeed, the broad range of CAC scores^27–31 among patients with both normal and abnormal perfusion scans in multiple studies, supports the concept that presence of calcified coronary atherosclerosis does not necessarily predict ischemia. Moreover, in subjects without overt CAD, the degree of CAC showed no relation³⁵ or a weak inverse relation to peak hyperemic myocardial blood flow and coronary flow reserve and direct relation to coronary vascular resistance.^36–38 The results of these studies suggest that CAC score and coronary microvascular function may provide biologically different information and could be complementary for risk assessment.

When combined with SPECT or PET MPI, CAC score provides independent and complementary information about risk of death or myocardial infarction.^30,31 In a previous SPECT study, including a lower-risk cohort with some asymptomatic subjects, and normal MPI, calcium scores did not show an incremental prognostic value over MPI over a mean follow-up of 32 months.³⁹ In contrast, with a longer follow-up (mean follow-up of 6.9 years), Chang et al.³⁰ demonstrated that individuals with high calcium score had greater annualized cardiac event rates (3%), despite a normal MPI (Fig. 25-5). The addition of a CAC score to ⁸²Rb perfusion data provided incremental prognostic information in patients with both ischemic and nonischemic perfusion studies.⁴⁰ These combined data supports the concept that while a normal relative MPI may indicate excellent short-term prognosis a high CAC score may indicate a worse intermediate-term prognosis despite a normal MPI.

Hybrid Coronary CTA and MPI

Imaging coronary atherosclerosis and its functional consequences with hybrid SPECT/CT and PET/CT devices in a single or a sequential study is now a reality. When performed on the same day, coronary CTA is typically performed after completion of stress MPI, so that beta-blockers can be administered to slow the heart rate for the coronary CTA without affecting ischemia assessment on MPI. Radionuclide and coronary CTA images can be interpreted independently or together using one of several software options to fuse the images.

The relationship between anatomic assessment of CAD and functional assessment of perfusion is complex. Lesion severity as assessed by coronary angiography does not account for the degree of endothelial dysfunction or the effect of serial stenosis on vascular resistance. Multiple single-center studies with coronary CTA and MPI have demonstrated the relatively poor ability of coronary CTA to predict ischemia on perfusion imaging with a modest PPV ~30%.⁴¹ Similarly, a normal myocardial perfusion scan is a relatively poor discriminator for the presence or absence of nonobstructive CAD.⁴² In addition, the extent of CAD can often be underestimated due to reduced heterogeneity of flow in patients with underlying CAD and concomitant endothelial dysfunction. Given the imperfect relationship between anatomic lesion severity and the degree of ischemia, a hybrid approach of radionuclide MPI and coronary CTA may allow for a more comprehensive characterization of CAD burden. However, both coronary CTA and MPI may not be indicated in all patients. A strategy of sequential imaging with initial MPI followed by coronary CTA (if MPI is not normal or severely abnormal) may be considered in subjects with an intermediate to high pretest likelihood. In contrast, an initial coronary CTA followed by MPI (unless the coronary CTA demonstrates normal arteries or critical CAD) may be a better strategy in subjects with low or low intermediate pretest likelihood of CAD. Others have proposed a coronary CTA along with stress MPI. An example of how this hybrid approach can provide useful clinical information is demonstrated in Figure 25-6.

Several studies to date support the notion that both CAC score and coronary CTA offer incremental diagnostic and prognostic information to MPI. A few clinical scenarios wherein a hybrid approach of combined MPI along with coronary CTA may be helpful are: (1) to detect severe multivessel CAD in the setting of mild perfusion abnormalities when the clinical suspicion is high (balanced ischemia); (2) to diagnose microvascular dysfunction in subjects with abnormal MPI by excluding atherosclerosis; (3) to evaluate patients with structural abnormalities of the coronary arteries.⁴³ Discordant findings on MPI and coronary CTA can result from microvascular dysfunction (abnormal blood flow without obstructive epicardial CAD, calcified and nonobstructive CAD with normal perfusion or obstructive CAD that is not flow-limiting (due to hemodynamic or collateral changes), and imaging artifacts. Therefore, combined MPI and CTA can provide better characterization of the extent and severity of underlying CAD and potential benefit from revascularization than does either technique alone.¹⁸ Indeed, one recent study showed that combining stress-only SPECT with coronary CTA in individuals presenting to the emergency room offers the added advantage of feasibility (no contraindications to SPECT), lower radiation dose when stress-only MPI was used, and higher prognostic value.⁴⁴ However, the coronary CTA approach was less costly, and none of the individuals with a 0 calcium score had significant CAD or cardiac event during follow-up.

There is also emerging evidence that diagnostic performance of SPECT or PET MPI and coronary CTA is superior than SPECT/PET MPI or coronary CTA alone (Table 25-3).^45–50 The EVINCI study, a multicenter study, included 292 symptomatic individuals with at least intermediate pretest likelihood of CAD, who underwent coronary CTA and at least one form of ischemia testing and were referred to invasive coronary angiography with an intention of evaluating FFR in intermediate lesions. These patients were followed up for 30 days and coronary revascularization was documented. Invasive coronary angiograms by QCA were considered obstructive with >50% stenosis in the left main or >70% stenosis in any of the other coronary arteries or 30% to 70% stenosis with an FFR ≤0.8. Majority of the individuals (70%) in the EVINCI study underwent SPECT MPI and the rest underwent PET MPI. Overall about 41% of the patients had normal hybrid imaging (MPI and coronary CTA normal) and 24% had a hybrid match (perfusion defect in the territory of a stenotic artery). As in prior single-center studies,^51–54 rate of coronary revascularization was highest in the matched group (70%), intermediate in the mismatch group (36%), and least in the normal group (10%), p < 0.001. However, radiation dose with the hybrid imaging approach was high (PET/coronary CTA: 9.4 mSv and SPECT/coronary CTA: 18.5 mSv (range 6–31 mSv). Before advocating a combined imaging approach, larger studies with longer-term follow-up are necessary to determine the optimal strategy for hybrid imaging. Whether a combined imaging strategy improves patient outcomes by identification of patients who will benefit from revascularization, avoidance of invasive angiography, or improved adherence to optimal medical therapy is still being determined.

Hybrid Cardiac CT and Radionuclide Imaging

Cardiac Inflammation and Infection

Hybrid imaging is uniquely suited to localize radiotracer uptake in hotspot radionuclide imaging and for molecular imaging (Fig. 25-7). As discussed in Chapter 24,¹⁸F-FDG imaging is emerging as a valuable technique to diagnose and manage sarcoidosis and prosthetic valve or cardiac device infection, and provides incremental value to diagnose infective endocarditis in prosthetic valves and intracardiac devices wherein echocardiography, cardiac CT, and cardiac magnetic resonance imaging (CMR) may be inconclusive.

Complex Congenital Heart Disease

Hybrid imaging is well suited for imaging individuals with complex congenital heart disease.^55,56 In addition to improved image quality, and attenuation correct, the CT portion of the PET or SPECT MPI helps distinguish prosthetic material-related perfusion defects from fibrosis.

Increased utilization of ionizing radiation in medical imaging, in large part attributable to CT scans and nuclear medicine studies, has led to increased scrutiny on the risk of radiation exposure from medical imaging.⁵⁷ Therefore, a chapter on hybrid SPECT–CT and PET–CT imaging must include a brief discussion on radiation risk and exposure, which is more completely addressed in Chapters 2 and 7. The estimated effective radiation doses for cardiac imaging studies are: coronary calcium CT 3 mSv, coronary CTA 2 to 18 mSv, ^99mTc-sestamibi 9 mSv, ⁸²rubidium PET 5 mSv, and CT for attenuation correction 0.2 mSv.³¹ The addition of a CAC score CT or a coronary CTA to an MPI study will further increase radiation dose.⁵⁷ The risk of this radiation exposure must be counterbalanced against the incremental diagnostic and prognostic information gained from the hybrid approach. Recently, measuring the cellular effects of radiation on DNA has emerged as an additional method of assessing radiation risk.^58,59 More research is needed to develop new agents to protect patients from the potential adverse effects of radiation and to decrease the radiation exposure.

Hybrid SPECT–CT and PET–CT systems offer attenuation-corrected MPI that improves the image quality and diagnostic accuracy compared to non–attenuation-corrected MPI. In addition to attenuation correction, the CT component of hybrid scanners can provide an accurate measure of calcified atherosclerotic burden and additional diagnostic and prognostic information to guide risk factor management, particularly when MPI is normal. Incorporation of coronary CTA with relative stress MPI and quantitative myocardial blood flow imaging is providing an intricate anatomic and physiologic assessment of CAD in each individual patient and offers the potential to guide personalized management of CAD. Hybrid PET–CT imaging with ¹⁸F-FDG PET–CT is breakthrough for imaging cardiac sarcoidosis, prosthetic valve and intracardiac device endocarditis, and molecular cardiology research. The ultimate transition to hybrid imaging with stress MPI and coronary CT angiography in clinical practice will be further defined and clarified with large, carefully designed and conducted clinical trials.