CT-Angiography Beats SPECT Perfusion Imaging for Coronary Disease Diagnosis in CORE320
October 23, 2015
RELATED LINKS
Use of FFR-CT Reduces Costs Compared With Diagnostic Angiography
DNA and Cellular Damage Observed Following Cardiac CT Scans
CT Can Do Work of SPECT and an Angiogram: CORE320
My Alerts
Click the topic below to receive emails when new articles are available.
BALTIMORE, MD — Computed-tomography angiography (CTA) is more accurate than nuclear-stress myocardial perfusion imaging (MPI) for the diagnosis of angiography-confirmed coronary artery disease in asymptomatic patients, according to the results of a new study[1].
Overall, CTA was more sensitive for detecting coronary artery disease in 391 asymptomatic patients participating in the CORE320 study, resulting in “greater overall accuracy,” say investigators. The area under the receiver operating characteristic curve (AUC) was 0.91 with CTA vs 0.69 for single-photon-emission CT-acquired MPI (SPECT-MPI), a statistically significant difference (P<0.001).
“CT coronary angiography yielded considerably greater accuracy than SPECT-MPI for establishing or excluding the diagnosis of angiographic coronary artery disease in symptomatic patients defined by the gold standard of cardiac catheterization,” write Dr Armin Arbab-Zadeh (Johns Hopkins University, Baltimore, MD) and colleagues October 8, 2015 in Circulation: Cardiovascular Imaging.
Overall Diagnostic Accuracy
| Measurement | CTA | SPECT | P |
| Area under the curve (AUC) | 0.91 | 0.69 | <0.001 |
| Sensitivity | 0.92 | 0.62 | <0.001 |
| Specificity | 0.75 | 0.68 | 0.23 |
| Positive predictive value (PPV) | 0.84 | 0.74 | 0.001 |
| Negative predictive value (NPV) | 0.60 | 0.55 | <0.0001 |
In their paper, Arbab-Zadeh, along with senior investigator Dr Julie Miller (Johns Hopkins University), note that stress testing, commonly combined with SPECT-MPI, often functions as a gatekeeper for patients with anginalike symptoms because of the risks and costs of taking the patient directly to the catheterization laboratory. CTA, they point out, is a noninvasive imaging test that has performed well, yielding high diagnostic accuracy, when compared with cardiac catheterization in several studies.
Overall, the researchers say the use of CTA was more accurate than SPECT-MPI for ruling in and ruling out coronary disease—confirmed by quantitative angiography—for all patients except those with severe coronary calcification and those with anatomy indicative of high-risk coronary disease. In patients with severe calcification, for example, diagnostic accuracy was similar between CTA and SPECT-MPI, a result that was mainly driven by less specificity for diagnosing coronary artery disease with CTA in patients with a calcium score >400. Similarly, in the 111 high-risk patients, specificity was also significantly lower than with SPECT-MPI.
The researchers say the performance of CTA was most “balanced” when a coronary stenosis threshold of 55% was used to identify patients with coronary artery disease on angiography. In that setting, the sensitivity and specificity was 88% and 83%, respectively, compared with 92% and 75%, respectively, when the prespecified 50%-stenosis threshold was used.
The mean radiation dose was CTA was 3.9 mSv, which was significantly lower than the 9.8 mSv recorded for patients undergoing SPECT-MPI.
The researchers conclude by stating that while CTA may be more effective than SPECT-MPI for diagnosing coronary disease, as well as ruling it out, recent trial results suggest screening with CTA is associated with more downstream cardiac catheterization and revascularization procedures, all without conclusive evidence of improved outcomes. Given this, “functional evaluation may be necessary before considering coronary interventions if CTA is used as first-line testing for coronary artery disease evaluation,” state Arbab-Zadeh and colleagues.
This work was funded in part by the division of intramural research of the National Heart, Lung, and Blood Institute, National Institutes of Health. Arbab-Zadeh and Miller served on the CORE320 steering committee. Disclosures for the coauthors are listed in the paper.