This evidence review was originally based on a literature search of the MEDLINE® (via PubMed) database through February 2004 and has been updated with subsequent literature review and/or repeat TEC Assessments.^9-11 The most recent literature review covers the period through October 20, 2015.

The objective of the 2005 TEC Assessment was to evaluate the clinical effectiveness of contrastenhanced cardiac computed tomography angiography (CTA) using either electron beam computed tomography (EBCT) or multidetector-row computed tomography (MDCT) as a noninvasive alternative to invasive coronary angiography (ICA), particularly in patients with a low probability of significant coronary artery stenosis. Evaluation of the coronary artery anatomy and morphology was the most frequent use of cardiac CTA and primary focus of the TEC Assessment. The Assessment considered multiple indications, but computed tomography (CT) technology used in studies reviewed is now outdated (studies employed 16-slice scanners). The TEC Assessment concluded that the use of contrast-enhanced cardiac CTA for screening or diagnostic evaluation of the coronary arteries did not meet TEC criteria.

The 2006 TEC Assessment was undertaken to determine the usefulness of cardiac CTA as a substitute for ICA for 2 indications: in the diagnosis of coronary artery stenosis and in the evaluation of acute chest pain in the emergency department (ED). Seven studies in the ambulatory setting and utilizing 40- to 64slice scanners were identified. Two studies performed in the ED used 4- or 16-slice scanners. Evidence was judged insufficient to form conclusions. Available studies at the time were inadequate to determine the effect of cardiac CTA on health outcomes for the diagnosis of coronary artery stenosis in patients referred for angiography or for evaluation of acute chest pain in the ED.

Three major indications for cardiac or coronary CTA are considered in the current evidence review: (1) patients with acute chest pain without known coronary disease presenting in the ED setting, (2) evaluation of stable patients with signs and symptoms of CAD in the non-ED setting, and (3) evaluation of anomalous coronary arteries.

Patients With Acute Chest Pain Presenting to the Emergency Setting

Diagnostic Validity

The diagnostic characteristics of coronary CTA have not been directly assessed in patients in the emergency setting. Because patients who test negative on CTA are discharged from care and the status of their disease is unknown, there is verification bias and diagnostic characteristics of CTA cannot be determined. The diagnostic characteristics of coronary CTA previously established in other studies was assumed to apply to patients in the ED setting and were tested in randomized trials to establish clinical utility.

Clinical Utility

A 2011 TEC Assessment examined evidence on the evaluation of patients with acute chest pain and without known coronary artery disease (CAD).¹¹ Randomized controlled trials (RCTs) and prospective observational studies were identified by searching the MEDLINE database and relevant bibliographies of key studies. Several RCTs of CTA conducted in emergency settings were identified.

A 2007 RCT by Goldstein et al evaluated 197 randomized patients from a single center without evidence of acute coronary syndromes to coronary CTA with 64-slice scanners (n=99) or usual care (n=98).¹² Over a 6-month follow-up, no cardiac events occurred in either arm. ICA rates were somewhat higher in the coronary CTA arm (12.1% vs 7.1%). Diagnosis was achieved more quickly after coronary CTA. 

A 2009 RCT evaluated a similar sample of 699 randomized patients from 16 centers: 361 undergoing coronary CTA with 64- to 320-slice scanners and 338 undergoing myocardial perfusion imaging (MPI).¹³ Over a 6-month follow-up, there were no deaths in either arm, 2 cardiac events in the coronary CTA arm and 1 in the perfusion imaging arm. ICA rates were similar in both arms (7.2% after coronary CTA, 6.5% after perfusion imaging). A second noninvasive test was obtained more often after coronary CTA (10.2% vs 2.1%), but cumulative radiation exposure in the coronary CTA arm (using retrospective gating) was significantly lower (mean 11.5 vs 12.8 mSv). Time to diagnosis was shorter (mean, 3.3 hours) and estimated ED costs lower with coronary CTA.

A 2012 RCT by Litt et al also evaluated the safety of coronary CT in the evaluation of patients in the ED.¹⁴

Although the study was a randomized comparison with traditional care, principal outcome was safety after negative CTA examinations. No patients who had negative CTA examinations (n=460) died or had a myocardial infarction (MI) within 30 days. Compared with traditional care, patients in the CTA group had higher rates of discharge from the ED (49.6% vs 22.7%), a shorter length of stay (median, 18.0 hours vs 24.8 hours), and a higher rate of detection of coronary disease (9.0% vs 3.5%). Three studies reported no cardiac events after a negative coronary CTA in the ED after 12- (N=481),¹⁵ 24- (N=368),¹⁶ or 47-month (N=506)¹⁷ follow-up.

A 2012 RCT by Hoffmann et al compared length of stay and patient outcomes in patients evaluated with CTA versus usual care.¹⁸ For patients in the CTA arm of the trial, mean length of hospital stay was reduced by 7.6 hours, and more patients were discharged directly from the ED (47% vs 12%). There were no undetected coronary syndromes and no differences in adverse events at 28 days. However, in the CTA arm, there was more subsequent diagnostic testing and higher cumulative radiation exposure. Cumulative costs of care were similar between the 2 groups.

A 2014 RCT by Hamilton-Craig et al compared length of stay and patient costs in 562 patients presenting to the ED with low-to-intermediate risk chest pain with CTA versus exercise stress testing.¹⁹ Costs within 30 days of presentation were significantly lower in the CTA group than in the exercise testing group (mean, $2193 vs $2704 p<0.001). Length of stay was significantly reduced in the CTA patients compared with the exercise testing group (mean, 13.5 hours vs 20.7 hours, p<0.0001). Clinical outcomes at 30 days and at 12 months did not differ.

Section Summary: Acute Chest Pain Presenting to the Emergency Setting

The high NPV of coronary CTA in patients presenting to the ED with chest pain allows coronary disease to be ruled out with high accuracy. The efficiency of the workup is improved, as patients are safely and quickly discharged from the ED with no adverse outcomes among patients who have negative CTA examinations.

Other important outcomes that require consideration in comparing technologies include ICA rates, use of a second noninvasive test, radiation exposure, and follow-up of any incidental findings. Although there is uncertainty accompanying the limited trial evidence, it is reasonable to conclude that ICA rate after coronary CTA is not markedly different from that after noninvasive imaging. Two studies showed that subsequent diagnostic testing was more frequent in patients who received CTA. Studies have differed in which treatment strategy results in higher overall radiation exposure. Incidental findings after coronary CTA are common and lead to further testing, but the impact of these findings on subsequent health outcomes is uncertain.

Stable Patients With Angina and Suspected CAD

Before the introduction of coronary CTA, the initial noninvasive test in a diagnostic strategy was always a functional test. Current practice guidelines recommend a noninvasive test be performed in patients with intermediate risk of coronary artery disease. The choice of functional test is based on clinical factors such the predicted risk of disease, electrocardiogram interpretability, and ability to exercise. When disease is detected, treatment alternatives include medical therapy or revascularization (percutaneous coronary intervention or coronary artery bypass graft surgery). If revascularization is indicated, patients undergo ICA to confirm the presence of stenosis. Which approach to adopt is based on the extent of anatomic disease, symptom severity, evidence of ischemia from functional testing, and more recently, fractional flow reserve obtained during invasive angiography. Many studies have shown that only a subset of anatomically defined coronary lesions are clinically significant and benefit from revascularization. Other studies have shown only limited benefits of treating coronary stenoses in stable patients. Thus an assessment of the diagnostic characteristics of coronary CTA alone is insufficient to establish clinical utility. A difficulty in evaluating a noninvasive diagnostic test for CAD is that the patient outcomes depend not only on the test results, but the management and treatment strategy. The most convincing evidence of clinical utility compares outcomes after anatomic-first (coronary CTA) and functional-first (eg, perfusion imaging, stress echocardiography) strategies.

Relevant studies reviewed here include studies comparing diagnostic performance of coronary CTA with angiography, studies of outcomes of patients undergoing CTA versus alternative tests, and studies of incidental findings and radiation exposure.

Diagnostic Accuracy

At this time, there is a fairly large body of evidence evaluating the diagnostic characteristics of coronary CTA for identifying coronary lesions. The best estimate of the diagnostic characteristics of coronary CTA can be obtained from review of recent meta-analyses and systematic reviews. Table 1 shows ranges of sensitivity and specificity for functional noninvasive tests from studies reviewed for the diagnosis and management of stable angina by Fihn et al.²⁰ Sensitivities tended to range between 70% and 90%, depending on the test and study, and specificities ranged between 70% and 90%.

For coronary CTA, estimates of sensitivity from various systematic reviews are considerably higher (see

Table 2). The guideline statement from Fihn et al cited studies reporting sensitivities between 93% and 97%.²⁰ A meta-analysis by Ollendorf et al of 42 studies showed a summary sensitivity estimate of 98% and a specificity of 85%. A meta-analysis of 8 studies conducted by the Ontario Health Ministry showed a summary sensitivity estimate of 97.7% and a specificity of 79%. In the meta-analysis by Nielsen et al, sensitivity of coronary CTA varied between 98% and 99% (depending on the analysis group).²¹

Table 1: Summary of Estimates of Sensitivity and Specificity of Functional Noninvasive Tests From Recent Guideline Statement (Fihn et al²⁰)

     Noninvasive Test                                           Sensitivity (Range or                Specificity (Range or

                                                                                                   Single Estimates)                      Single Estimates)

Exercise electrocardiography		61%	70%-77%
Pharmacologic stress echocardiography		85%-90%	79%-90%
Exercise stress echocardiography		70%-85%	77%-89%
Exercise myocardial perfusion imaging		82%-88%	70%-88%
Pharmacologic stress myocardial perfusion imaging		88%-91%	75%-90%

Table 2: Estimates of Sensitivity and Specificity of Coronary Computed Tomography Angiography From Guidelines and Meta-Analyses 

               Study                                                      Sensitivity (Range or      Specificity (Range or

                                                                                                     Single Estimates)           Single Estimates)

20

Fihn et al (2012) guideline statement	93%-97%	80%-90%
Ollendorf et al (2011) meta-analysis22	98%	85%
Health Quality Ontario (2010) meta-analysis23	97.7%	79%
Nielsen et al (2014) meta-analysis21	98%-99%	82%-88%

Clinical Utility

Randomized Controlled Trials

 

For patients at intermediate risk of CAD, 5 RCTs were identified comparing net health outcomes following a coronary CTA strategy with outcomes from other noninvasive testing strategies.

The PROMISE trial randomized 10,003 patients to coronary CTA or exercise electrocardiography, nuclear stress testing, or stress echocardiography (as determined by physician preference) as the initial diagnostic evaluation.²⁴ For the composite end point of death, MI, hospitalization for unstable angina, or major procedural complication, the outcome rate between the 2 groups showed no statistically significant difference (hazard ratio [HR], 1.04; 95% confidence interval [CI], 0.83 to 1.29). Coronary CTA also did not meet prespecified noninferiority criteria compared with alternative testing. Some clinical outcomes assessed at 12 months favored coronary CTA, but the differences were nonsignificant. Coronary catheterization rates and revascularization rates were higher in the coronary CTA group.

In the SCOT-HEART trial, 4146 patients were randomized to coronary CTA or standard care. The primary end point was the change in the proportion of patients with a more certain diagnosis (presence or absence) of angina pectoris.²⁵ Secondary outcomes included death, MI, revascularization procedures, and hospitalizations for chest pain. Analysis of the primary outcome showed that patients who underwent coronary CTA had an increase in the certainty of their diagnosis relative to those in usual care (relative risk, 1.79; 95% CI, 1.62 to 1.96). Regarding health outcomes, the rate of heart disease death and MI was lower with coronary CTA (1.3% vs 2.0%; HR=0.62; p=0.053), but results were of marginal statistical significance.

A small trial by Min et al (2012) randomized 180 patients with stable chest pain to initial diagnostic evaluation using coronary CTA or MPI.²⁶ The primary outcome was angina-specific health status. There were no significant differences at follow-up between groups on any measures of health status.

The FACTOR-64 trial randomized 900 subjects with diabetes to screening with coronary CTA or standard care.²⁷ Patients in this trial were asymptomatic, but were considered to be at high risk for CAD due to long-standing diabetes. The primary outcome was a composite of mortality, nonfatal MI, or unstable angina requiring hospitalization. At a median follow-up of 4 years, there was no significant difference between the groups for the primary outcome (CTA, 6.2%; control, 7.6%; HR=0.80; p=0.38).

The CAPP trial randomized 500 patients with stable chest pain to coronary CTA or exercise stress testing.²⁸ The primary outcome was the change difference in scores of Seattle Angina Questionnaire domains at 3 months. Patients were also followed up for further diagnostic tests and management. In the CTA arm, 15.2% of subjects underwent revascularization. In the exercise stress testing arm, 7.7% underwent revascularization. For the primary outcome, angina stability and quality of life showed significantly greater improvement in the coronary CTA arm than in the exercise stress testing arm. 

Nonrandomized Studies

 

Four nonrandomized studies were identified comparing outcomes of patients following a coronary CTA strategy with outcomes following other noninvasive testing strategies. A nonrandomized study of coronary CTA with computed fractional flow reserve assessment is discussed in the Appendix.²⁹ Some studies emphasized downstream utilization of diagnostic testing and procedures rather than patient outcomes.

Nielsen et al conducted an observational trial comparing patients who underwent coronary CTA and exercise stress testing.³⁰ Patients had a low-to-intermediate pretest probability of CAD and presented with suspected angina. Patients were followed for 12 months after the initial test, and assessed for occurrence of major adverse events such as cardiac death and nonfatal MI. Subsequent utilization of cardiovascular tests and therapy was also compared between groups. Clinical outcomes were not compared formally because there were few clinical events. No deaths were reported during the follow-up period. Three patients in the exercise testing group had MIs within 12 months. For downstream test utilization, the exercise test group had greater subsequent use perfusion imaging (9% vs 4%, p=0.03) and greater mean total 1-year costs (€1777 vs €1510, p=0.03). Rates of ICA and revascularization did not differ statistically significantly.

Shreibati et al used Medicare claims data to compare all-cause mortality, subsequent utilization of several cardiac tests, treatment, and total costs in patients who underwent initial noninvasive testing with either coronary CTA, stress echocardiography, MPI, or exercise electrocardiography.³¹ In this study, patients undergoing coronary CTA had higher rates of several types of utilization subsequent to their test than patients undergoing MPI. The study also presented outcomes for both stress echocardiography and exercise electrocardiography, but they tended not to be any different than for MPI. There were increased rates of ICA (22.9% vs 12.1%) and revascularization (11.4% vs 4.6%). Total spending and CAD-related spending were also higher for coronary CTA than for MPI. There was no significant difference in all-cause mortality between coronary CTA and MPI. Although the mortality rate for coronary CTA was slightly lower than the mortality rate for MPI (1.05% vs 1.28%), the adjusted odds ratio (OR) showed a higher risk of mortality, which may be due to unusual confounding. However, there was a slightly lower likelihood of hospitalization for MI (adjusted OR=0.60; p=0.04).

In Min et al (2008), costs and clinical outcomes for patients undergoing initial coronary CTA were compared with patients undergoing initial MPI.³² The data source for this study was a proprietary claims database from 2 regional health plans. Utilization of medical care was lower after coronary CTA. Overall costs were lower, the proportion receiving ICA was lower, and the proportion receiving revascularization was lower after coronary CTA. In terms of clinical outcomes, the proportion with a hospitalization for angina was lower in the coronary CTA group. The coronary CTA group also had a lower rate of a combined outcome of angina or MI hospitalization (HR=0.70; 95% CI, 0.55 to 0.90).

In 2825 patients evaluated for stable angina and suspected CAD in Japan, Yamauchi et al examined outcomes after initial coronary CTA (n=625), MPI (n=1205), or angiography (n=950).³³ Average follow-up was 1.4 years. In a Cox proportional hazards model adjusted for potential confounders, the relative hazard of major cardiac events after MPI or coronary CTA were lower than after angiography; annual rates were 2.6%, 2.1%, and 7.0%, respectively. Revascularization rates were higher after coronary CTA than MPI (OR=1.6; 95% CI, 1.2 to 2.2).

Incidental Findings

Nine studies using 64+-slice scanners were identified.^34-42 Incidental findings were frequent (26.6%-

68.7%) with pulmonary nodules typically the most common and cancers rare (?5/1000 or less). Aglan et al (2010)³⁴ compared the prevalence of incidental findings when the field of view was narrowly confined to the cardiac structures with that when the entire thorax was imaged. As expected, incidental findings were less frequent in the restricted field (clinically significant findings in 14% vs 24% when the entire field was imaged).

Radiation Exposure

Exposure to ionizing radiation increases lifetime cancer risk.⁴³ Three studies have estimated excess cancer risks due to radiation exposure from coronary CTA.^6,7,44 Assuming a 16-mSv dose, Berrington de Gonzalez et al (2009)⁴⁴ estimated that the 2.6 million coronary CTAs performed in 2007 would result in 2700 cancers or approximately 1 per 1000. Smith-Bindman et al (2009) estimated that cancer would develop in 1 of 270 women and 1 of 600 men age 40 undergoing coronary CTA with a 22-mSv dose.⁷ Einstein et al (2007) employed a standardized phantom to estimate organ dose from 64-slice coronary CTA.⁶ With modulation and exposures of 15 mSv in men and 19 mSv in women, calculated lifetime cancer risk at age 40 was 7 per 1000 men (1/143) and 23 per 1000 women (1/43). However, estimated radiation exposure used in these studies was considerably higher than received with current scanners?now typically under 10 mSv and often less than 5 mSv with contemporary machines and radiation reduction techniques. For example, in the 47-center PROTECTION I study enrolling 685 patients, the mean radiation dose was 3.6 mSv, using a sequential scanning technique.⁴⁵ In a 2012 study of patients undergoing an axial scanning protocol, mean radiation dose was 3.5 mSv, and produced equivalent ratings of image quality compared with helical scan protocols, which had much higher mean radiation doses of 11.2 mSv.⁴⁶ 

Section Summary: Stable Angina and Suspected CAD

A number of studies have evaluated the diagnostic accuracy of CTA for diagnosing CAD in an outpatient population. In general, these studies report high sensitivity and specificity, although there is some variability in these parameters across studies. Meta-analysis of these studies showed that for detection of anatomic disease, coronary CTA has a sensitivity greater than 95%, which is superior to all other functional noninvasive tests. Specificity is at least as good as other noninvasive tests. However, the link between improved diagnosis and health outcomes is not as clear, and thus outcome studies are necessary to demonstrate the clinical utility of coronary CTA.

Direct clinical trial evidence comparing coronary CTA and other strategies in the diagnostic management of stable patients with suspected CAD has not demonstrated the superiority of coronary CTA in any of the single clinical trials. Clinical trials demonstrated greater utilization of ICA and subsequent revascularization procedures after coronary CTA. An important problem of interpreting the clinical trials is that the comparator strategies differ: in the PROMISE trial, the CAPP trial, and Min et al (2012), coronary CTA was compared with an alternative noninvasive test; in other studies, coronary CTA was supplement to usual care (which may or may not have included a noninvasive test). This design difference in the clinical trials is likely a reflection of how coronary CTA is used in clinical practice?either as a substitute for another noninvasive test or as an addition to other noninvasive tests. The PROMISE trial explicitly compared coronary CTA with an alternative functional test as the initial diagnostic test. Although the trial did not show the superiority of coronary CTA and did not meet prespecified criteria for noninferiority, Examination of some secondary clinical outcomes supports a conclusion of “at least” noninferiority. The results of the other randomized trials are consistent with noninferiority of coronary CTA with other established noninvasive tests. Thus, the randomized studies indicate that outcomes of patients are likely to be similar with coronary CTA versus other noninvasive tests.

 The nonrandomized studies of coronary CTA have several methodologic shortcomings including reliance on administrative data and inability to fully assess and adjust for potential confounding. The findings generally show little difference in patient outcomes between diagnostic strategies. Downstream utilization of medical care showed variable findings.

Although studies of incidental findings and radiation exposure raise issues regarding the potential for adverse effects of coronary CTA, there is not sufficient evidence that the magnitude of these effects is important for ascertaining the net benefit or risk of coronary CTA in this setting.

Anomalous Coronary Arteries

Anomalous coronary arteries are an uncommon finding during angiography, occurring in approximately 1% of coronary angiograms completed for evaluation of chest pain. However, these congenital anomalies can be clinically important depending on the course of the anomalous arteries. A number of case series have consistently reported that coronary CTA is able to delineate the course of these anomalous arteries, even when conventional angiography cannot.^47-50 However, none of the studies reported results when the initial reason for the study was to identify these anomalies, nor did any of the studies discuss impact on therapeutic decisions. Given the uncommon occurrence of these symptomatic anomalies, it is unlikely that a prospective trial of coronary CTA could be completed. 

Other Diagnostic Uses of Coronary CTA

Given its ability to define coronary artery anatomy, there are many other potential diagnostic uses of coronary CTA including patency of coronary artery bypass grafts, in-stent restenosis, screening, and preoperative evaluation.

•       Evaluating patency of vein grafts is generally less of a technical challenge due to vein size and lesser motion during imaging. In contrast, internal mammary grafts may be more difficult to image due to their small size and presence of surgical clips. Finally, assessing native vessels distal to grafts presents difficulties, especially when calcifications are present, due to their small size. For example, a 2008 meta-analysis including results from 64-slice scanners, reported high sensitivity 98% (95% CI, 95 to 99; 740 segments) and specificity 97% (95% CI, 94 to 97).⁵¹ Other small studies have reported high sensitivity and specificity.^52,53 Lacking are multicenter studies demonstrating likely clinical benefit, particularly given the reasonably high disease prevalence in patients evaluated.

•       Use of coronary CTA for evaluation of in-stent restenosis presents other technical challenges?motion, beam hardening, and partial volume averaging. Whether these challenges can be sufficiently overcome to obtain sufficient accuracy and impact outcomes has not been demonstrated.

•       Use for screening a low-risk population was recently evaluated in 1000 patients undergoing coronary CTA compared with a control group of 1000 similar patients.⁵⁴ Findings were abnormal in 215 screened patients. Over 18 months of follow-up, screening was associated with more invasive testing, statin use, but without difference in cardiac event rates.  

Coronary CTA for preoperative evaluation before noncardiac surgery has been suggested, but evaluated only in small studies and lacking demonstrable clinical benefit.

Ongoing and Unpublished Clinical Trials

Some currently unpublished trials that might influence this review are listed in Table 3.

Table 3. Summary of Key Trials 

NCT Number	Title	Enrollment	Completion Date
Ongoing
NCT01384448	Stress Echocardiography and Heart Computed Tomography (CT)	400	Jun 2015

Scan in Emergency Department Patients With Chest Pain

NCT01283659	IMAGE-HF Project I-C: Computed Tomographic Coronary Angiography for Heart Failure Patients (CTA-HF)	250	Jun 2016
NCT01083134	The Correlation of Heart Hemodynamic Status Between 320 Multidetector Computed Tomography, Echocardiography and Cardiac Catheterization in Patients With Coronary Artery Disease	100	Mar 2020
NCT01559467	The Supplementary Role of Non-invasive Imaging to Routine Clinical Practice in Suspected Non-ST-elevation Myocardial Infarction (CARMENTA)	300	Apr 2015
NCT02400229	Diagnostic Imaging Strategies for Patients With Stable Chest Pain and Intermediate Risk of Coronary Artery Disease (DISCHARGE	3546	Sep 2019
NCT02291484	Comprehensive Cardiac CT Versus Exercise Testing in Suspected Coronary Artery Disease (2) (CRESCENT2)	250	Dec 2015
Unpublished
NCT00991835	Plaque Registration and Event Detection In Computed Tomography (PREDICT)	3015	Dec 2014

NCT: national clinical trial.

Summary of Evidence

The evidence for coronary CTA in patients who present with chest pain and suspected coronary artery disease in the emergency setting, at intermediate to low risk, includes several randomized controlled trials. Relevant outcomes are overall survival, morbid events, and resource utilization. The studies showed similar patient outcomes, with faster patient discharges from the emergency department, and lower short-term costs. The evidence is sufficient to determine qualitatively that the technology results in a meaningful improvement in the net health outcome.

The evidence for coronary CTA in stable patients who have chest pain, intermediate risk of coronary artery disease, and meet guideline criteria for requiring noninvasive testing (ie intermediate risk) includes studies of diagnostic accuracy of coronary CTA, randomized trials comparing coronary CTA with alternative diagnostic strategies, and observational studies comparing coronary CTA with alternative diagnostic strategies. Relevant outcomes are overall survival, test accuracy, morbid events, and resource utilization. Studies of diagnostic accuracy show that coronary CTA has higher sensitivity and similar specificity to alternative noninvasive tests. Although randomized trials do not show superiority of coronary CTA to other diagnostic strategies, they are consistent with noninferiority (ie, similar health outcomes) to other diagnostic strategies. The evidence is sufficient to determine qualitatively that the technology results in a meaningful improvement in the net health outcome.

The evidence for coronary CTA in patients who have a clinical scenario where anomalous coronary arteries are suspected includes case series. Relevant outcomes are overall survival, test accuracy, morbid events, and resource utilization. The studies show that coronary CTA can often detect anomalous coronary arteries that are missed by other diagnostic modalities. Anomalous coronary arteries are a rare condition, and formal studies to assess clinical utility are unlikely to be performed. In most situations, these studies would be insufficient to determine whether the test improves health outcomes. However, for this rare condition, evidence is sufficient to determine qualitatively that the technology results in a meaningful improvement in the net health outcome.