OUP user menu

CCTA to guide revascularization for high-risk CAD: a ‘cliff hanger’

Pim J. de Feyter , Koen Nieman
DOI: http://dx.doi.org/10.1093/eurheartj/ehs371 3011-3013 First published online: 1 November 2012

This editorial refers to ‘All-cause mortality benefit of coronary revascularization vs. medical therapy in patients without known coronary artery disease undergoing computed tomographic angiography: results from CONFIRM (COronary CT Angiography EvaluatioN For Clinical Outcomes: An InteRnational Multicenter Registry)’, by J.K. Min et al., on page 3088

From its inception in 1958, invasive coronary angiography (ICA) has been the sole key imaging technique to diagnose coronary obstructions and to guide revascularization, initially by bypass graft surgery (CABG), and later by balloon angioplasty and stent implantation. However, ICA comes at a price, with, inherent to the invasive nature of the procedure, patient inconvenience, complications (albeit low), and costs. To contain referral to ICA, clinicans consider the pre-test probability of obstructive coronary artery disease (CAD) and employ non-invasive stress tests to assess provocable myocardial ischaemia as a gatekeeper to ICA. The pre-test probability can be estimated using algorithms based on age, gender, symptom characteristics, traditional risk factors of cardiovascular disease, and the resting electrocardiogram (ECG). Numerous non-invasive stress tests are available to assess the presence and functional severity of CAD, which is associated with clinical outcome, and guide further decision making. Patients without evidence of stress-induced myocardial ischaemia have an excellent prognosis and generally do not need further testing. Patients with minimal to moderate ischaemia may initially be treated medically, reserving ICA for those with symptoms refractory to optimal medical therapy. Patients with severe ischaemia are usually directly referred for ICA to determine suitability for revascularization with percutaneous coronary intervention (PCI) or CABG.

Numerous studies have convincingly shown that revascularization of high-risk angiographic CAD is associated with better clinical outcome than medical therapy.1 Angiographic high-risk CAD is defined as two-vessel disease (VD) involving the left anterior descending (LAD), three-VD, or left main. However, current guidelines recommend revascularization of these patients only in the presence of objective myocardial ischaemia by invasive or non-invasive means.2 These recommendations are based on non-randomized observational data, including a study by Hachamovitch et al.3 of 10 627 consecutive patients who underwent stress testing with a follow-up of 1.9 ± 0.6 years, that demonstrated that revascularization compared with medical treatment had a greater survival benefit in patients with moderate to severe ischaemia, whereas patients with no or mild ischaemia fared better with medical treatment. Despite the established overall prognostic value of stress imaging, it remains worrisome that a substantial number of patients with angiographic high-risk CAD for various technical reasons remain undetected with non-invasive functional testing and do not benefit from revascularization.4,5 The value of non-invasive coronary computed tomography angiography (CCTA) to exclude obstructive CAD, which is associated with an excellent prognosis, is widely recognized. If CCTA would also be able to differentiate high-risk patients who would prognostically benefit from revascularization based on the extent of angiographic disease, that could have considerable consequences for the future diagnostic management of patients with suspected CAD.

In the new study by Min and colleagues, the correlation between CCTA findings and outcome after revascularization was investigated to assess the potential value of CCTA to guide revascularization decisions.6 From the CONFIRM (COronary CT Angiography EvaluatioN For Clinical Outcomes: An InteRnational Multicenter Registry) registry they studied 15 223 stable patients without known CAD who underwent ≥64-detector row CCTA for suspected CAD. The patients were followed for a median of 2.1 years. All-cause mortality was the clinical endpoint, which occurred in 1.2% (185) of the patients. High-risk CAD was defined as two-VD ( ≥70%) with involvement of the proximal LAD, three-VD, or left main CAD ( ≥50%), which was diagnosed by CCTA in 1103 patients (7.3%). Post-CCTA revascularization, defined as having occurred within 90 days after CCTA, was performed in 569 patients (5.2%). A propensity score was developed to account for non-randomized referral to revascularization. Revascularization was performed in 51.2% of high-risk patients and in 3.8% of non-high-risk patients. The all-cause mortality rate in high-risk CAD patients was higher in medically treated patients compared with the revascularized group (5.34 vs. 2.28, P = 0.075), while the opposite was observed in non-high-risk patients: 0.97% vs. 2.06% (P = 0.0138). In a multivariate model, revascularization compared with medical treatment was associated with a lower all-cause mortality risk [hazard ratio (HR) 0.38, 95% confidence interval (CI) 0.18–0.83] in high-risk CAD patients, whereas no difference was observed in non-high-risk CAD patients (HR 3.24, 95% CI 0.76–13.9).

The study concludes that in the present day, with improved medical treatment of stable CAD, angiographic severity assessed by a non-invasive technique predicts benefit from revascularization and may be used to guide referral to revascularization. There are, however, several aspects that need to be considered while interpreting these intriguing results.

  1. Despite the size of this multicentre registry and the sophisticated statistical analysis, this observational, non-randomized study remains vulnerable to selection bias. While the ultimate association between CCTA findings and benefit from revascularization is evident, it remains unclear which clinical and diagnostic parameters determined therapeutic management. The fact that a patient with severe CAD is deferred from revascularization may be due to more subtle differences in disease severity or (prognostially relevant) co-morbidities. In this respect, instead of all-cause mortality as the endpoint, differentiation between cardiac and non-cardiac mortality would, despite its acknowledged limitations, have been more representative of the clinical benefit.

  2. The observation that ∼30% of the total population that underwent CCTA and 24% of the revascularized group had no symptoms, while 13% of revascularized patients had no angiographically obstructive CAD on CCTA, is remarkable. The subgroup analysis shows that the absent, or negative, effect of revascularization was particularly profound in patients without symptoms.

  3. Information regarding functional testing before or after CCTA is not reported. It remains unclear to what extent invasive angiography and/or revascularization was driven by functional or anatomical findings, and what their respective value is towards clinical benefit.

  4. Data on invasive angiographic (or functional) findings, and the accuracy of CCTA to predict high-risk CAD by the reference standard are unavailable. It may be assumed that patients underwent revascularization according to ICA outcome and not solely according to CCTA outcome. This may have introduced selection of only patients with ‘true’ high-risk angiographic CAD in whom revascularization is associated with the highest benefit.

  5. There was a lack of knowledge of medical treatment during follow-up which may have affected clinical outcome.

  6. Will surgeons accept the accuracy of CCTA to plan which vessels to graft?

Nevertheless the authors are to be commended for demonstrating for the first time that the angiographic severity of CAD by CCTA predicts clinical outcome after revascularization, and recognizing the potential role of this non-invasive modality in selecting patients who would benefit from these therapeutic procedures. The authors are prudent and critical by clearly stating that the findings of this non-randomized observational study should be considered hypothesis generating. The study is exciting and provocative, but at this point in time should be regarded as a ‘cliff hanger’ study: we are kept in suspense and will have to wait for further research to reveal how the story will end.

First-line testing with non-invasive coronary computed tomography angiography: a game changer?

Non-invasive functional testing is the established first-line approach to triage patients with stable angina to medical treatment or revascularization, but may miss patients with angiographic high-risk CAD known to benefit from revascularization. As an alternative first-line anatomical test, CCTA may be used because it has a high diagnostic accuracy to rule out or rule in obstructive CAD.7 However CCTA cannot reliably predict the functional significance of coronary obstructions, in particular intermediate obstructions. CCTA is able to identify angiographic high-risk CAD with the caveat that not all anatomic high-risk lesions are flow limiting. A substantial number of coronary obstructions in these angiographic high-risk CAD patients appear, after assessment with invasive fractional flow reserve determination, not to be flow limiting. This may result in reclassification of these high-risk CAD patients into moderate risk CAD or even low-risk patients in whom functional revascularization or medical treatment is preferred.8

A diagnostic algorithm that complements the merits of both functional and anatomical testing may be more effective to triage patients to optimal medical treatment or optimal medical treatment and revascularization.

A suggested algorithm is provided in Figure 1. The preference for an anatomical test as first-line diagnostic work-up instead of a functional test is based on the favourable outcome of the FAME 1 and 2, the only randomized trials with clinical outcome showing that medical treatment or revascularization should be guided first with anatomy and subsequently with function.9,10

Figure 1

Diagnostic algorithm for patients with stable angina. CAD, coronary artery disease; CCTA, coronary computer tomography angiography; CA, coronary angiography; Rx, treatment; FFR, fractional flow reserve.

Conflict of interest: none declared.


  • The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.