The decision scheme on the diagnostic work-up for CTO patients to evaluate eligibility for revascularization is depicted in Fig. 1. Please note that this algorithm is based on current international guidelines and is not specifically targeted to CTO patients but to obstructive CAD in general. Once the diagnosis of a CTO has been established, usually through a diagnostic invasive coronary angiogram (ICA), information pertaining LV function is mandatory. Normal regional LV function of myocardium subtended by the CTO by definition excludes non-viability. The decision to proceed to a revascularization procedure is then merely based on the presence and extent of ischemia. Without documented ischemia, such a procedure is futile and even hazardous [30] as the patient is subjected to potential procedural complications without a clear benefit and OMT suffices. In the majority of cases, however, ischemia is observed. Even the presence of well-developed collaterals does generally not protect the myocardium against ischemia. An increase in collateral flow reserve during pharmacological stress has been shown to occur in only 7 % of patients, while coronary steal can be documented in a third [31, 32]. When the jeopardized myocardium comprises more than 10 % of the LV, revascularization on top of OMT is considered appropriate irrespective of symptoms [33]. Retrospective analysis of myocardial perfusion studies in large registries have revealed that such an ischemic burden holds prognostic relevance and revascularization is associated with improved outcome [34]. When the extent of ischemia is less pronounced (i.e., < 10 %), an initial treatment strategy of OMT is justified and revascularization only contemplated after inadequate response to medical therapy.

Conversely, when patients exhibit extensive regional wall motion abnormalities in the affected vascular territory, we recommend that the diagnostic work-up focus on myocardial viability rather than ischemia. Without signs of myocardial viability (i.e., previous (near) transmural myocardial infarction), similar to lack of ischemia, revascularization is considered inappropriate as no recovery of function is anticipated [35, 36]. Although, it must be noted that revascularization of non-viable myocardium might halt the process of progressive LV remodelling [37]. The presence of substantial viability warrants revascularization and is accompanied by recovery of LV function and improved outcome as observed in both retrospective and prospective studies [35, 38, 39]. The extent of viability deemed necessary for a clinically relevant improvement in LV ejection fraction (i.e., >5 percentage points) is uncertain, but generally ranges from 10 to 25 % jeopardized myocardium [40, 41]. Small quantities of viability are not associated with recovery of function, but revascularization could be considered to reduce symptoms not controlled by medical therapy.

Although the presented literature supports the utilization of the proposed treatment algorithm, randomized trials on imaging-guided revascularization are scarce. The PET and Recovery Following Revascularization (PARR-2) [42] and a substudy of the Surgical Treatment for Ischemic Heart Failure (STICH) [43] trial have evaluated the impact of viability assessment on clinical outcomes after CABG and were both negative. A nuclear substudy of the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial suggests that an appreciable reduction of ischemia is accompanied by a favorable outcome after PCI [44]. These results were corroborated for CTO patients who underwent PCI, whereby a baseline ischemic burden >12.5 % in combination with a reduction >5 % yielded the most favorable clinical outcomes [45]. A randomized imaging study on the merits of ischemia-guided revascularization is, however, lacking altogether. The ongoing International Study of Comparative Health Effectiveness with Medical and Invasive Approaches (ISCHEMIA) trial is attempting to establish the value of revascularization in patients with a large ischemic burden in a randomized fashion. More specifically tailored to PCI-CTO, large observational studies have shown that a successful percutaneous recanalization results in improved all-cause mortality, lower MACE rates (relative risk: 0.70), and reduced need for CABG as opposed to unsuccessful attempts [13, 18, 19, 46]. Nonetheless, these studies are all observational in nature and thus prone to confounding. Currently, the Drug-Eluting Stent Implantation Versus Optimal Medical Treatment in Patients with Chronic Total Occlusion (DECISION-CTO) and the Randomized Multicentre Trial to Evaluate the Utilization of Revascularization or Optimal Medical Therapy for the Treatment of Chronic Total Coronary Occlusions (EURO-CTO) trials are the first studies to more accurately assess the surplus clinical value of PCI-CTO over OMT alone in a randomized fashion. The results of the aforementioned studies are eagerly awaited.

Although a large armamentarium of imaging modalities is available to evaluate myocardial perfusion and viability, PET is unequivocally accepted as the reference standard for this task. The following section will place the value of PET within the context of alternative imaging techniques.

Over the last decade, CCTA has developed as a valuable non-invasive alternative for the visualization of coronary anatomy. Current multislice CT scanners in combination with modern acquisition protocols enable robust and reproducible assessment of coronary artery morphology with relatively high temporal and spatial resolution accomplished at an acceptable radiation dose [65‐68]. CCTA allows to visualize the characteristics of a CTO, which is useful not only for diagnostic purposes but also for planning a potential interventional strategy. CCTA even has some advantages over invasive angiography with three-dimensional reconstructions of the affected segment to fully appreciate the course and length of the occlusion (Fig. 8).

A CTO on CCTA is diagnosed if there is complete discontinuity of contrast opacification of the coronary artery lumen in the cross-sectional views, multiplanar reconstructions (MPR), and maximum intensity projections (MIP). The time lapse between the intravenous contrast injection and the contrast bolus tracking setoff results in less dense and reverse contrast opacification distal to the occlusion due to collateral flow [69, 70]. Suspicion of a more recent occluded vessel should rise if there is no distal contrast opacification [71, 72]. Von Erffa et al. showed that an occlusion length of ≥9 mm on CCTA predicted a total occlusion on ICA with 100 % specificity in a group of 40 consecutive patients with a lesion demonstrating complete contrast discontinuity on CCTA. Figure 8 shows the accuracy of occlusion length on CCTA in comparison with ICA. Nevertheless, differentiation between high-grade stenosis and total occlusion remains unreliable due to the relatively limited spatial resolution of CCTA [73], while the presence of true lumen antegrade flow is of clinically importance since it significantly influences the success rate of PCI.

Historically, the success rate of PCI CTO ranged between 50 and 70 %, which has resulted in the reluctance to attempt percutaneous revascularization [1, 7, 12]. To more adequately predict the chances of successful antegrade wire crossing (within 30 min) for appropriate case selection, the Japanese Multicenter CTO Registry (J-CTO) score was developed [28•]. The J-CTO score has identified several independent angiographic predictors of failure being heavy calcification, bending within the occluded segment, blunt proximal stump, and occlusion length >20 mm. An additional point in this scoring system is given for a previous failed attempt. Low J-CTO scores are thus easily crossed, while high scores are characterized by high likelihood of failure. One of the issues in the full assessment of the J-CTO score is the frequent necessity of simultaneous double arterial injection of the right and left coronary arteries (see Figs. 2, 7, 8, and 9). These images are generally not produced during a diagnostic invasive angiogram. These angiographic features, however, can also be visualized by CCTA and could therefore similarly aid in the prediction of procedural success (Table 2). Several small-scaled studies have indeed identified very comparable angiographic CTO characteristics obtained with CCTA that were associated with PCI CTO failure as those observed in the J-CTO registry [74‐78]. It has even been suggested that pre-procedural knowledge of the three-dimensional anatomy of the occlusion may aid in antegrade wire escalation and crossing of the lesion and thus improve procedural outcome [79]. Figure 9 illustrates that CCTA can be quite helpful to appreciate the coronary course of an occluded segment. More recently, Opolski et al. [80••] reported a CCTA-based scoring system in 240 CTO patients to predict success of percutaneous antegrade wire crossing in the Computed Tomography Registry of Chronic Total Occlusion Revascularization (CT-RECTOR). Very much like the J-CTO, independent angiographic predictors were occlusion length >20 mm, multiple occlusions, blunt stump, bending, and severe calcification. Clinical predictors were previously failed attempt and CTO duration >12 months or unknown duration. The predictive value of the CT-RECTOR proved to be superior over the J-CTO score to predict antegrade wire crossing <30 min (AUC 0.83 vs. 0.71). Figure 10 shows an example of a low and high CTO difficulty score on CCTA and ICA. These data suggest that CCTA could be utilized in clinical practice for case selection and perhaps even for pre-procedural planning to improve success. There is, however, an important caveat in the interpretation of these data. All of these studies have merely investigated the likelihood of antegrade wire crossing. Even though this type of approach is still the mainstay in attempts to percutaneously tackle CTOs, its success rate is low (62 % in the CT-RECTOR Registry). Dedicated CTO centers that have adopted the “hybrid” approach (i.e., next to antegrade wire escalation implementing retrograde wire escalation and dissection re-entry techniques) nowadays achieve success rates over 90 % with very acceptable complication rates. This approach (as depicted in Fig. 11) lets coronary anatomy define the procedural strategy, but anatomical features and lesion complexity are not used to select cases [29••]. As such, CCTA is hardly utilized by hybrid operators. Nonetheless, pre-procedural CCTA could be useful for antegrade wire operators. Based on non-invasive CT imaging, lesion complexity and thus chances of success can be accurately assessed and this information employed to select cases or refer to expert centers [81].