The Year in Cardiac CT Angiography: Highlights For the General Cardiology Fellow
Dec 8, 2015 | Bryan J. LeBude, MD
Since the advent of electron-beam computed tomography (CT) in 1990, great technical advances have enabled the development of various applications useful in evaluating the cardiovascular patient. In the current era, coronary computed tomography angiography (CCTA) provides excellent image quality at continually decreasing doses of radiation. Research in the field has shifted away from confirming diagnostic accuracy towards demonstrating incremental benefit over existing approaches. The past year has provided further clinical trial evidence to guide both the evaluation of the stable patient with suspected coronary artery disease (CAD), and the high-risk asymptomatic patient:
CCTA screening of asymptomatic patients with diabetes fails to significantly decrease mortality and cardiac events compared with aggressive medical therapy (FACTOR-64)
Rationale/Design: In light of the fact that diabetes is strongly associated with CAD, in addition to worse outcomes following revascularization, early identification is a reasonable target for improving outcomes. The FACTOR-64 trial randomized 900 asymptomatic patients with diabetes to screening CCTA or medical therapy over a median follow-up of four years. Targets for the control group were consistent with guidelines-based care including target A1C < 7 percent, LDL < 100 and SBP < 130 mmHg. Recommendations for intensification of medical therapy or coronary angiography were given to the CCTA group based on imaging.
Results: The primary outcome (morality, nonfatal myocardial infarction (MI) or unstable angina requiring hospitalization) was not significantly different between the CCTA and control group (6.2 percent vs. 7.6 percent, P = 0.38). Ischemic major adverse cardiovascular events also did not differ between CCTA and control (4.4 percent vs. 3.8 percent, P = 0.68). Results are consistent with previous trials such as DIAD and BARI-2D. Various reasons have been cited for the lack of observed benefit. Excellent baseline medical therapy was achieved, as evidenced by the mean LDL-C (87.7 mg/dL) and SBP (130.5 mmHg) in the control group. Intensified medical therapy in the CCTA group also did not significantly increase risk factor reduction, and these patients only had a modestly increased rate of revascularization (5.8 percent). Overall, there was a low prevalence of severe (>70 percent) proximal stenosis (6.3 percent).
In patients with symptoms concerning for underlying CAD, anatomical testing with CCTA yields similar clinical outcomes compared with functional stress testing (PROMISE)
Rationale/Design: For symptomatic patients without known CAD, there is little consensus as to the optimal approach to noninvasive testing. This void has led to the testing of very low risk populations, and high rates of non-obstructive CAD found on catheterization. PROMISE randomized 10,003 patients with symptoms suspicious for CAD to CCTA or functional testing with stress nuclear (67 percent), stress echo (23 percent) or exercise ECG (10 percent). The patient population was at intermediate risk of CAD with a 53.3 percent calculated pretest likelihood of obstructive disease.
Results: At a median follow-up of 25 months, there was no difference in the primary endpoint (3.3 percent vs. 3 percent, P = 0.75) which included all-cause mortality, MI, unstable angina hospitalization or major complication. More patients in the CCTA group underwent catheterization (12.2 percent vs. 8.1 percent) or revascularization (6.2 percent vs. 3.2 percent), with a lower proportion found to have non-obstructive CAD (3.4 percent vs. 4.3 percent) on catheterization. Median cumulative radiation exposure was lower in CCTA (10.0 mSV) vs. functional (11.3 mSV), but with a higher mean (12.0 vs 10.1 mSV). Subsequent economic analysis showed a small, non-significant increase in cumulative costs for CCTA vs. functional testing. Critics have noted several limitations of the trial. Power to detect differences in outcomes was limited by a much lower observed rate of the primary endpoint compared with the expected 8 percent. In addition, there was no evidence that anatomical testing led to changes in medical therapy; because imaging can only influence outcomes by changing care, ineffective treatments or the inability to follow best practices can separate downstream outcomes from the initial test.
CCTA increases diagnostic certainty and alters planned treatment regimens for patients presenting with stable chest pain (SCOT-HEART)
Rationale/Design: SCOT-HEART randomized 4146 outpatients with stable chest pain to CCTA or standard of care, which most often included stress electrocardiogram testing (85 percent). The primary objective was to investigate how CCTA alters diagnostic certainty, with secondary objectives of assessing the influence on downstream management of these patients and how this relates to outcomes.
Results: The primary endpoint of the study, certainty that the symptoms are attributable to CAD, increased significantly (HR 1.79, CI 1.62-1.96) for the CCTA group. The six-week diagnosis of angina due to CAD changed in 23 percent of the CCTA participants versus 1 percent of standard care. Changes in planned investigations (15 percent vs. 1 percent, P<0.0001) and recommendations for preventive treatment (18 percent vs. 4 percent, P<0.0001) were also significantly higher in the CCTA group. Despite no differences in angina, revascularization or hospital admission, there was a non-significant decrease in the rate of CAD death and MI (1.3 percent vs. 2.0 percent, P = 0.0527) with CCTA. This signal must be interpreted with caution, as it was small in absolute number and it was one of 22 prespecified secondary endpoints.
Overall, CCTA is a safe and effective alternative to functional stress testing for the evaluation of patients with suspected CAD. Additional work is needed to fully delineate the comparative value of cardiovascular imaging in this patient population. The demonstration of differences in downstream clinical outcomes among testing modalities will require a large clinical trial that maintains a close link between testing results and patient management.
This article was authored by Bryan J. LeBude, MD, a fellow in training at MedStar Washington Hospital Center in Washington, DC.
Acknowledgement: I would like to thank Ana Barac, MD, PhD, FACC, for her thoughtful article review.