with less frequent 2‐stent overlap and less stenting of the side branch in another study [26]. CTA was even shown to outperform the current Medina classification and RESOLVE angiographic scoring system in predicting intraprocedural side branch occlusion [27].
Stents
Coronary stents are challenging to evaluate on CTA because of partial voluming artifact that causes apparent enlargement of the stent and obscures the lumen (Figure 10.1c). This is exaggerated with smaller stents and overlapping or bifurcation stents.
To evaluate for in‐stent restenosis (ISR), the in‐stent lumen must be directly visualized (Figure 10.1b). Contrast enhancement distal to the stent could be secondary to retrograde filling by collaterals. Improvement in CTA technology has led to an improvement in evaluation of ISR. A recent meta‐analysis evaluated 35 studies with data from 2656 patients that used CTA to diagnose ISR. Not only did CTA have a high sensitivity and specificity for detection of ISR but also by demonstrating a high positive and low negative Likelihood Ratio, the authors concluded that CTA was an excellent test to rule in and rule out ISR, and serve as a good screening test. CTA was more sensitive for stents with a diameter of >3mm, more accurate for stents with thickness <100microns and less accurate at higher heart rates [28].
CTA has been shown to be reliable in excluding ISR in stents implanted in the left main (LM) and proximal left anterior descending/circumflex (LAD/Cx), as these stents have larger lumens [29]. This is important because routine angiographic surveillance 6 months after PCI for LM STEMI is recommended [30] and CTA is considered appropriate for this purpose [31].
Chronic total occlusion (CTO)
Successful revascularization of CTOs continues to be one of the more challenging procedures in cardiology. CTA has the advantage of better characterizing certain key features such as tortuosity of vessels, ostial lesions, and length of occluded segments. Several studies have demonstrated that specific CT findings and CT‐based scores can predict failure of CTO PCI [32]. While more data is required, CTA has the potential to play a key role in planning CTO revascularization procedures in the coming years.
Coronary artery bypass grafting (CABG)
CTA has been shown to be accurate in evaluating venous and arterial bypass grafts because they have a larger diameter and little calcification (Figure 10.1c–d). The challenge lies in evaluating the native coronaries, which are typically severely calcified in post‐CABG patients. Jungman et al. [33] reviewed 13 studies that assessed bypass grafts for stenosis of more than 50% and found that the sensitivity and specificity of CTA approached nearly 100%. This included two studies that used 265‐slice and 320‐slice CT, which were found to have similar accuracy to 64‐slice CT. The advantage of the newer scanners was the reduction in radiation dose. The study did reiterate though, that the sensitivity and specificity for evaluating native coronaries was lower. Currently, it is considered appropriate to use CTA for evaluation of graft patency in patients with ischemic symptoms [24]. Interestingly, a recent study showed that in patients with left main or triple‐vessel disease, the heart team management strategy based on CTA alone was in agreement with that made by ICA – suggesting that future decisions regarding revascularization strategy could possibly be based on CTA findings alone [34].
Trials and current guidelines
CTA has been proven to have high NPV and can be used safely to rule out acute coronary syndrome (ACS) to expedite discharge in low to intermediate risk patients presenting with acute chest pain [1] [35–40]. The use of CTA as the initial screening test for CAD has remained somewhat controversial due to its tendency to overestimate disease severity. Two large trials of symptomatic patients referred for CAD evaluation are the PROMISE and the SCOT‐HEART trials [41, 42]. The PROMISE trial randomized symptomatic patients to CTA testing or functional testing and found no significant difference in clinical outcomes (composite of death, MI hospitalization for unstable angina or major procedural complication) over a median period of two years. The SCOTHEART trial randomized patients to standard care plus CCTA or standard care alone and at 5 years, found a significantly lower rate of death from coronary heart disease or myocardial infarction in the CCTA group. There was an initial increase in the rate of procedures in the CCTA group but there was no difference between the two groups at the end of five years. The authors concluded that since CTA was shown to be more accurate in diagnosing coronary obstruction in both SCOTHEART and PROMISE, their results implied that PCIs were more appropriately performed in the CCTA group. Additionally, SCOTHEART encouraged initiation of secondary prevention strategies in patients with non‐obstructive disease and the proportion of patients receiving this was higher in the CTA group because of lack of recognition in the usual care group. The 5‐year outcomes of the SCOTHEART trial are very encouraging with regards to the use of CCTA in the assessment of stable ischemic heart disease. On account of its high sensitivity and cost‐effectiveness, the updated 2016 NICE guidelines [43] now recommend the use of coronary CT as the initial screening test for all patients with stable angina. Additionally, a recent trial randomized patients presenting with NSTEMI to an initial ICA, CMR or CTA strategy and found that while fewer procedures occurred in the latter two groups with unchanged one‐year clinical outcomes – implying that CTA or CMR could possibly serve as gatekeepers to ICA in low‐risk NSTEMI as well [44].
CT FFR
Fractional Flow Reserve (FFR) on ICA is the gold standard for hemodynamic assessment of coronary lesions. CT FFR is a technology that uses computational fluid dynamics to calculate FFR non‐invasively. It provides a combination of anatomic and functional information that helps identify high‐risk plaque and accurately determine the severity of coronary stenosis. Several studies have evaluated the accuracy of CT FFR. The NXT trial [45] found that CT FFR calculated on 254 patients with predominantly intermediate stenosis (30–70%) had high accuracy compared to invasive FFR and that the specificity of CT FFR was higher than CCTA alone. In the PLATFORM study [46], symptomatic patients with intermediate pre‐test probability were divided into non‐invasive and invasive groups and consecutive cohorts in each group were evaluated with usual care or CT FFR. In the invasive group, CT FFR led to a significant reduction in costs with a similar rate of revascularization and change in Quality of Life (QOL) scores. In the non‐invasive group, the CT FFR cohort had a significantly higher improvement in QOL scores. With the incorporation of CT FFR in routine testing, we can expect to see the use of coronary CT continue to expand.
TAVR
Over the last decade, TAVR has revolutionized the management of aortic stenosis. The aortic and mitral valves are well visualized on CTA [1], which is why CTA is now considered the imaging gold standard for pre‐TAVR evaluation and is increasingly being used for pre‐TMVR evaluation as well. A pre‐TAVR CT is a comprehensive study. The thoracic portion of the contrast‐enhanced study is ECG‐gated and must cover the aortic root to allow for appropriate measurements of the aortic annulus, sinuses of Valsalva, heights of the coronary ostia from the annulus, sinotubular junction and fluoroscopic angle. The abdominal and pelvic portion is used to measure iliofemoral and subclavian vessel size and is not ECG‐gated. Cardiac CT also plays a significant role in predicting new persistent left bundle branch block (NP‐LBBB) post‐TAVR. Two recent studies evaluated the CT‐based predictors of NP‐LBBB in patients undergoing TAVR using an Edwards Sapien 3 (S3) or EVOLUT R/PRO valve. The first study found that advanced age, a higher NTI, or Non tubular index (lower NTI is an indicator of a more tubular LVOT), more annular calcification on the right coronary cusp, implant depth at the non coronary cusp and LVOT oversizing >10% are independent predictors of NP‐LBBB when the
