Coronary CTA has been consistently applied for the prediction of time efficiency (i.e. guidewire crossing within 30 min) and the procedural outcome of CTO PCI (i.e. guidewire crossing with restoration of thrombolysis in myocardial infarction grade 3 flow). To this end, several single CTO features as well as combined scoring systems based on CTA have been used to assess the difficulty level of the CTO prior to PCI. Notably, all of the trials on the role of CTA for the prediction of the procedural outcome of CTO PCI were limited by their observational design, and mainly focused on antegrade wiring strategy [2, 14‐25].

The shape of the proximal cap on angiography is paramount for planning antegrade wiring—whereas a tapered stump indicates the presence of ‘microchannels’ that are usually negotiated with soft polymer-jacketed wires, a blunt stump necessitates the use of stiffer and more penetrating wires [26]. Although coronary CTA can differentiate between a blunt and tapered proximal cap (including the exact origin and direction of the side branches), CT studies have provided conflicting results concerning the influence of stump morphology on success rates in CTO PCI [2, 14, 16, 17, 21, 25]. These disparities may result from varying levels of operator experience and varying CTO PCI techniques across different clinical sites as well as a potentially high variability in the assessment of stump morphology on coronary CTA [25].

Apart from assessing the shape of the CTO entry, coronary CTA is uniquely suited for clear visualisation of the proximal cap relative to the further course of the occlusion site—a phenomenon not readily available by any other imaging modality. This virtually precludes the occurrence of an ambiguous cap on CTA, and can be potentially applied for proximal cap clarification in lesions with inconclusive angiographic findings—including guiding of proximal cap puncture (Fig. 1a–d), balloon-assisted subintimal entry (Fig. 2), and the scratch and go technique. In this context, coronary CTA can be also used for localisation of the stumps of the occluded bypass grafts and guiding their recanalisation for retrograde access to CTO PCI (Fig. 1e–h).

The extent and severity of calcifications are proportional to the age of coronary occlusions, and increase the difficulty of CTO PCI [6]. Noteworthy is that CTA is more sensitive in identifying, localising and quantifying calcification as compared with invasive angiography [15, 25]. Several CT studies have consistently demonstrated the independent relationship between calcification and procedural failure in CTO PCI [2, 14‐19, 21, 24, 25, 27‐29]. Specifically, calcification involving ≥50% of the vessel cross-sectional area (CSA) within the occlusion site (rather than calcification length) is considered the most frequent quantitative correlate of failed CTO PCI [2, 14, 15, 17, 21, 25, 27]. In addition, it was suggested that calcification at the proximal cap (but not at the distal cap) constitutes the major hurdle of successful guidewire crossing when using the antegrade approach [2, 21].

Hence, the identification of the extent and severity of calcification (preferably by using the ≥50% cut-off value for vessel CSA) on coronary CTA is paramount for guidewire and microcatheter selection, planning wiring escalation and de-escalation strategies, guiding the knuckle wire technique (Fig. 3), and marking the most optimal re-entry sites in the antegrade dissection and re-entry strategy.

CTO tortuosity has significant implications for appropriate guidewire selection. In CTO lesions with marked tortuosity (particularly over long distances), polymer-jacketed wires are preferred due to the lower risk for coronary perforation, whereas straight and short CTO segments can usually be traversed with stiffer wires [26]. Significantly, CTA enables straightforward recognition and precise quantification of coronary tortuosity, which is frequently underestimated on invasive angiography [5, 13]. Particularly germane to this concept, marked vessel tortuosity throughout the occluded segment (most commonly defined as an angle ≥45°) on CTA was repeatedly associated with procedural failure (although not always as an independent predictive factor) [16, 25, 27].

The recognition of marked CTO vessel tortuosity on non-invasive CTA can thus enhance appropriate guidewire selection, and lower the threshold for the dissection and re-entry strategy. In addition, the timely identification of pronounced CTO bending should prompt the operator to choose the optimal fluoroscopic angulation with the least foreshortening of the occlusion site based on three-dimensional CTA reconstructions (Fig. 3).

Angiographic occlusion length ≥ 20 mm has been accepted as a handy and approximate correlate of failed antegrade wiring [28]. Although coronary CTA surpasses invasive angiography in accurate measurement of the occlusion length, the relevance of CT-defined occlusion length in CTO PCI is uncertain [5, 13]. Indeed, while some CT studies indicated the independent association between long occlusions (defined as either >15 mm or >18 mm or ≥32 mm or as a continuous variable) and failed PCI [14, 18, 20, 23, 24, 27], others failed to confirm any independent relationship between occlusion length and CTO PCI outcome [2, 15‐17, 19, 21, 22, 25]. These disparities may reflect differences in the application of the hybrid CTO PCI techniques (mainly dissection and re-entry techniques) by different operators for resolving long occlusions.

Besides occlusion length, the relevance of multiple occlusion sites in significantly reducing the chances of time-efficient guidewire crossing has been reported [25]. Specifically, the detrimental effect of multiple occlusions on recanalisation attempts relies on longer occlusion length and multiple entry and exit sites, hampering successful guidewire passage into the distal true lumen. It is noteworthy that coronary CTA is superior to invasive angiography for visualisation of multiple occlusion sites [25].

To date three CTA scoring systems have been developed to predict the procedural outcome of CTO PCI, and thus to grade the CTO difficulty level prior to PCI (Tab. 1). Notably, the trials on CTA-derived scoring systems comprised European, Japanese, and Chinese populations enrolled between 2007 and 2016, and predominantly focused on antegrade wiring strategies [25, 27, 29, 30]. Hence, the clinical applicability of CT scores for prediction of CTO PCI success rates using retrograde wiring and dissection and re-entry strategies is largely unknown.

Originally, the CT-RECTOR (Computed Tomography Registry of Chronic Total Occlusion Revascularization) score was introduced as an easy-to-use prediction rule for 30-min guidewire crossing derived from 240 CTO lesions from a multicentre European registry [25]. The score features six independent predictors (multiple occlusions, blunt stump, severe calcification occupying ≥50% vessel CSA, bending, CTO age ≥12 months or unknown, previously failed CTO PCI), and is calculated by assigning 1 point for each of the variables and summing all points accrued (Fig. 4). Notably, the CT-RECTOR score was externally validated in a Chinese population as a more accurate prediction rule for 30-min guidewire crossing and final procedural success rate as compared with the angiographic J‑CTO (Multicenter Chronic Total Occlusion Registry of Japan) score [31]. Subsequently, two trials explored the predictive ability of the J‑CTO score derived from CTA (the so-called J‑CTO_CT score) [29, 30]. Interestingly, while in one study the non-invasive J‑CTO score determined by CTA yielded a similar predictive value for 30-min guidewire crossing as compared with the angiographic J‑CTO score [29], the other study demonstrated significantly higher discriminating accuracy of the J‑CTO_CT score than that of the angiography-based J‑CTO score [30]. Noteworthy is that the J‑CTO scores determined by coronary CTA and invasive angiography differ in the amount of surveyed calcification—while on CTA calcification must involve >50% of the vessel CSA, invasive angiography stratifies the presence or absence of calcification only. Finally, the KCCT (Korean Multicenter CTO CT Registry) score was developed in the derivation cohort of 456 CTO lesions, yielding seven independent predictors (proximal blunt entry, proximal side branch, bending, occlusion length ≥15 mm, severe calcification or whole luminal calcification, reattempt, and ≥12 months or unknown duration of occlusion) of 30-min guidewire crossing [27]. Whereas it showed higher discriminative performance compared with the other scoring systems, including J‑CTO, CL (clinical and lesion-related), PROGRESS-CTO (Prospective Global Registry for the Study of Chronic Total Occlusion Intervention), and CT-RECTOR scores (with the least numerical difference between KCCT and CT-RECTOR score), some of the parameters included in the KCCT score (e.g. proximal side branch, occlusion length ≥15 mm) may be less relevant when hybrid techniques (e.g. balloon-assisted subintimal entry) and modern CTO PCI devices are applied [26, 32].

In summary, clinicians may find the CTA-based scores particularly useful to better estimate the time and resources required for the interventional treatment of CTO, and to plan training sessions for less-skilled interventionalists.

Apparently, the biggest advantage of coronary CTA for facilitating CTO PCI relies on real-time CTA co-registration in the catheterisation laboratory [13]. To this end, one approach has been to import preprocedural CT datasets directly to the angiography system with subsequent display of CTA images on separate monitors. Specifically, such a solution features a fully automatic alignment of CT reconstructions according to the angulation of the C‑arm, enabling a continuous three-dimensional CTA roadmap during PCI [5, 13]. In addition, the colour-coded display of coronary arteries in virtual CTA facilitates contrast-free identification of fluoroscopic projections without foreshortening for each respective vessel segment [5]. This has been corroborated in a study by Rolf et al., who reported a significantly higher CTO recanalisation success rate in 25 patients with CTA co-registration as compared with 25 patients without CTA overlay [33]. Subsequently, the periprocedural co-registration of CTA using a wearable computer resulted in improved efficiency and less use of resources (as reflected by more frequent selection of the first-choice stiffer wire and lower contrast exposure), although there were no differences in CTO success rate as compared with standard CTO PCI [34]. Notably, both of these studies were limited by observational design and antegrade wiring strategies, so that the results need to be interpreted with caution [33, 34].

Recently, a more challenging approach involving real-time fusion of three-dimensional coronary CTA with X‑ray fluoroscopy for guiding CTO PCI has been investigated. In a study by Ghoshhajra et al. among 24 consecutive CTO PCI patients, CTA/fluoroscopy fusion provided new insights into the extent and localisation of coronary calcification as well as vessel tortuosity, potentially influencing antegrade wiring, antegrade and dissection re-entry, and retrograde wiring strategies [35]. Similar findings were reported by Xenogiannis et al., in whose study CTA/fluoroscopy fusion was used in 27 of 146 CTO PCI cases performed between 2018 and 2019 within the PROGRESS-CTO registry [36]. Noteworthy, the most common indication for CTA/fluoroscopy fusion was to clarify proximal cap ambiguity, followed by facilitation of intra-CTO wire advancement and re-entry. This subsequently contributed to antegrade and dissection re-entry being the most common successful crossing technique in the CTA/fluoroscopy fusion group—a phenomenon with a potential for reducing the need for a retrograde approach. Notably, although both of these studies were small and had an observational design, they are the first to examine the influence of coronary CTA for guiding CTO PCI using the hybrid algorithm [35, 36].

Finally, the interchangeable CTA scanning and analysis intertwined with invasive fluoroscopy has been tested in a series of 61 patients undergoing CTO PCI to investigate the role of intraprocedural CTA for identification of guidewire tip locations [37]. Whereas on-site acquisition and review of the CT images in the catheterisation laboratory was safe and feasible, not every CTA result provided precise information about guidewire location (not to mention potentially higher labour input and radiation dose of intraprocedural CTA).