Comparison of Coronary CT Angiography, SPECT, PET, and Hybrid Imaging for Diagnosis of Ischemic Heart Disease Determined by Fractional Flow Reserve

Abstract

Importance: At present, the choice of noninvasive testing for a diagnosis of significant coronary artery disease (CAD) is ambiguous, but nuclear myocardial perfusion imaging with single-photon emission tomography (SPECT) or positron emission tomography (PET) and coronary computed tomography angiography (CCTA) is predominantly used for this purpose. However, to date, prospective head-to-head studies are lacking regarding the diagnostic accuracy of these imaging modalities. Furthermore, the combination of anatomical and functional assessments configuring a hybrid approach may yield improved accuracy.

Objectives: To establish the diagnostic accuracy of CCTA, SPECT, and PET and explore the incremental value of hybrid imaging compared with fractional flow reserve.

Design, setting, and participants: A prospective clinical study involving 208 patients with suspected CAD who underwent CCTA, technetium 99m/tetrofosmin-labeled SPECT, and [15O]H2O PET with examination of all coronary arteries by fractional flow reserve was performed from January 23, 2012, to October 25, 2014. Scans were interpreted by core laboratories on an intention-to-diagnose basis. Hybrid images were generated in case of abnormal noninvasive anatomical or functional test results.

Main outcomes and measures: Hemodynamically significant stenosis in at least 1 coronary artery as indicated by a fractional flow reserve of 0.80 or less and relative diagnostic accuracy of SPECT, PET, and CCTA in detecting hemodynamically significant CAD.

Results: Of the 208 patients in the study (76 women and 132 men; mean [SD] age, 58 [9] years), 92 (44.2%) had significant CAD (fractional flow reserve ≤0.80). Sensitivity was 90% (95% CI, 82%-95%) for CCTA, 57% (95% CI, 46%-67%) for SPECT, and 87% (95% CI, 78%-93%) for PET, whereas specificity was 60% (95% CI, 51%-69%) for CCTA, 94% (95% CI, 88%-98%) for SPECT, and 84% (95% CI, 75%-89%) for PET. Single-photon emission tomography was found to be noninferior to PET in terms of specificity (P < .001) but not in terms of sensitivity (P > .99) using the predefined absolute margin of 10%. Diagnostic accuracy was highest for PET (85%; 95% CI, 80%-90%) compared with that of CCTA (74%; 95% CI, 67%-79%; P = .003) and SPECT (77%; 95% CI, 71%-83%; P = .02). Diagnostic accuracy was not enhanced by either hybrid SPECT and CCTA (76%; 95% CI, 70%-82%; P = .75) or by PET and CCTA (84%; 95% CI, 79%-89%; P = .82), but resulted in an increase in specificity (P = .004) at the cost of a decrease in sensitivity (P = .001).

Conclusions and relevance: This controlled clinical head-to-head comparative study revealed PET to exhibit the highest accuracy for diagnosis of myocardial ischemia. Furthermore, a combined anatomical and functional assessment does not add incremental diagnostic value but guides clinical decision-making in an unsalutary fashion.

Conflict of interest statement

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Leipsic reported having core laboratory contracts with Edwards Lifesciences, for which he receives no direct compensation. Dr Knuuti reported receiving support from the Academy of Finland Centre of Excellence in Molecular Imaging in Cardiovascular and Metabolic Research, Helsinki, Finland, and receiving grant support from Gilead Inc and serving as a consultant to Lantheus Inc. Dr Min reported serving as a consultant to HeartFlow and Abbott Vascular, serving on the scientific advisory board of Arineta, and holding an equity interest in MDDX. Dr van Royen reported receiving educational grants from Baxter and Biotronik. Dr Lammertsma reported receiving research grants from AVID, Philips Healthcare, F. Hoffmann–La Roche Ltd, and the European Commission. No other disclosures were reported.

Figures

Figure 1.. Schematic Illustration of the Study Protocol

All patients underwent the imaging protocol at day 1 (A) and day 2 (B) and were subsequently referred to the catheterization laboratory for invasive coronary angiography in conjunction with fractional flow reserve measurements. CAC indicates coronary artery calcium; CCTA, coronary computed tomography angiography; LD-CT, low-dose computed tomography; PET, positron emission tomography; and SPECT, single-photon emission computed tomography.

Figure 2.. Diagnostic Performance of Coronary Computed Tomography Angiography, Single-Photon Emission Computed Tomography, and Positron Emission Tomography Imaging for Diagnosis of Significant Coronary Artery Disease as Defined by Different Standards

FFR indicates fractional flow reserve; ICA, invasive coronary angiography; NPV, negative predictive value; and PPV, positive predictive value.

Figure 3.. Diagnostic Performance of Cardiac Imaging Methods and Traditional Cardiovascular Risk Factors for the Detection of Coronary Artery Disease (CAD) on a Patient-Based Level

Area under the curve (AUC) calculated for coronary computed tomography angiography (CCTA), single-photon emission computed tomography (SPECT), positron emission tomography (PET), and cardiovascular risk factors for diagnosing ischemia (PET AUC = 0.90 [95% CI, 0.85-0.94; P < .001]; CCTA AUC = 0.86 [95% CI, 0.81-0.91; P < .001]; SPECT AUC = 0.86 [95% CI, 0.81-0.91; P < .001]; and cardiovascular risk factors AUC = 0.77 [95% CI, 0.71-0.84; P < .001]). Comparison of diagnostic performances of noninvasive cardiac imaging and traditional risk factors for the detection of hemodynamically significant CAD was determined by receiver operating characteristic curves (PET vs CCTA: AUC = 0.90 vs 0.86 [P = .18]; PET vs SPECT: AUC = 0.90 vs 0.86 [P = .10]; SPECT vs CCTA: AUC = 0.86 vs 0.86 [P = .98]; PET vs cardiovascular risk factors: AUC = 0.90 vs 0.77 [P < .001]; SPECT vs cardiovascular risk factors: AUC = 0.86 vs 0.77 [P < .001]; and CCTA vs cardiovascular risk factors: AUC = 0.86 vs 0.77 [P < .001]).