Lesion characteristics and procedural complications of chronic total occlusion percutaneous coronary intervention in patients with prior bypass surgery: A meta‐analysis

Abstract Coronary artery bypass graft (CABG) accelerates the prevalence of native coronary chronic total occlusion (CTO), and this kind of CTO shows extensive challenging and complex atherosclerotic pathology. As a result, the procedural success rate of percutaneous coronary intervention (PCI) is inferior to another kind of lesions. The present meta‐analysis aims to compare the lesion characteristics and procedural complications of CTO‐PCI in patients with or without prior CABG. A total of 8 studies, comprising of 13439 patients, published from inception to August 2021 were included in this meta‐analysis. Results were pooled using random effects model and are presented as odds ratio (OR) with 95% confidence intervals (95% CIs). From the 13439 patients enrolled, 3349 (24.9%) patients had previous CABG and 10090 (75.1%) formed the control group in our analysis. For the clinical characteristic, compared to the non‐CABG patients, prior CABG patients were older (OR, 3.98; 95% CI, 3.19–4.78; p < .001; I 2 = 72%), had more male (OR, 1.30; 95% CI, 1.14–1.49; p < .001; I 2 = 6%), diabetes mellitus (OR, 1.54; 95% CI, 1.36–1.73; p < .001; I 2 = 37%), dyslipidemia (OR, 1.89; 95% CI, 1.33–2.69; p < .001; I 2 = 81%), hypertension (OR, 1.88; 95% CI, 1.46–2.41; p < .001; I 2 = 71%), previous myocardial infarction (OR, 1.94; 95% CI, 1.48–2.56; p < .001; I 2 = 85%), and previous PCI (OR, 1.74; 95% CI, 1.52–1.98; p < .001; I 2 = 22%). Non‐CABG patents had more current smoker (OR, .45; 95% CI, 0.27–0.74; p < .001; I 2 = 91%). BMI (OR, −0.01; 95% CI, −0.07–0.06; p = .85; I 2 = 36%) were similar in both groups. For lesions location, the right coronary artery (RCA) was predominant target vessel in both groups (50.5% vs 48.7%; p＝.49), although, the left circumflex (LCX) was more frequently CTO in the prior CABG group (27.3% vs 18.9%; p＜.01), while left anterior descending artery (LAD) in non‐CABG ones (16.0% vs 29.1%; p＜0.01). For lesions characteristics, prior CABG patients had more blunt stump (OR, 1.71; 95% CI, 1.46–2.00; p < .001; I 2 = 40%), proximal cap ambiguity (OR, 1.45; 95% CI, 1.28–1.64; p < .001; I 2 = 0.0%), severe calcifications (OR, 2.91; 95% CI, 2.19–3.86; p < .001; I 2 = 83%), more bending (OR, 3.07; 95% CI, 2.61–3.62; p < .001; I 2 = 0%), lesion length > 20 mm (OR, 1.59; 95% CI, 1.10–2.29; p = .01; I 2 = 83%), inadequate distal landing zone (OR, 1.95; 95% CI, 1.75–2.18; p＜.001; I 2 = 0.0%), distal cap at bifurcation (OR, 1.65; 95% CI, 1.46–1.88; p < .001; I 2 = 0.0%), and higher J‐CTO score (SMD, 0.52; 95% CI, 0.42–0.63; p < .001; I 2 = 65%). But side branch at proximal entry (OR, 0.88; 95% CI, 0.72–1.07; p = .21; I 2  = 45%), in‐stent CTO (OR, 0.99; 95% CI, 0.86–1.14; p = .88; I 2 = 0.0%), lack of interventional collaterals (OR, 0.80; 95% CI, 0.55–1.15; p = .23; I 2 = 78%), and previously failed attempt (OR, 0.73; 95% CI, 0.48–1.11; p = .14; I 2 = 89%) were similar in both groups. For complication, prior CABG patients had more perforation with need for intervention (OR, 1.91; 95% CI, 1.36–2.69; p < 0.001; I 2 = 34%), contrast‐induced nephropathy (OR, 3.40; 95% CI, 1.31–8.78; p = .01; I 2 = 0.0%). Non‐CABG patents had more tamponade (OR, 0.25; 95% CI, 0.09–0.72; p = .01; I 2 = 0.0%), and the major bleeding complication (OR, 1.18; 95% CI, 0.57–2.44; p = .65; I 2 = 0%) were no significant difference in both groups. In conclusion, Patients with prior CABG undergoing CTO‐PCI have more complex lesion characteristics, though procedural complication rates were comparable.

Although several experts have regarded previous CABG as a predictor of technical failure in CTO-PCI. 5 Patients who underwent previous CABG have exhibited more inflammation, fibrosis, calcification, and negative remodeling in CTO lesions compared with CABG-naïve patients for pathologic examination. 6 Consequently, previous CABG has been included as a risk factor for procedural technical failure in the Registry of CrossBoss and Hybrid procedures in France, the Netherlands, Belgium, and the United Kingdom (RE-CHARGE) score. 5 Over the past decade, thanks to the remarkable update of revascularization techniques, equipment, and contemporary algorithms in CTO-PCI procedures, CTO-PCI success rates have approached 90% for experienced operators currently. 7 Thus, PCI has emerged as a promising alternative treatment for CTO revascularization in post-CABG patients.
However, the lesion characteristics of CTO-PCI for a native coronary artery in previous CABG status have been sparsely studied, and it remains unclear whether the higher procedural difficulty encountered during CTO-PCI in patients who have undergone CABG is also mirrored by worse complications. Some recent studies investigated the lesion characteristics of CTO-PCI in previous CABG patients; however, the results were inconsistent. Furthermore, whether the higher procedural complexity encountered during CTO-PCI in previous CABG patients translated into major complications is yet to be clarified. Given the increasing prevalence of patients with CTO after CABG, along with the development of novel interventional approaches, characterization, and complications of these patients gained further importance. To shed further light on this issue, we conducted a systematic review and meta-analysis to assess clinical and procedural characteristics as well as in-hospital major complications of CTO-PCI in patients with and without CABG.

| Study selection
Studies were included when the following were satisfied: (1) studies with patients treated by CTO-PCI. CTO was defined as a TIMI grade 0 flow in coronary artery for more than 3 months; (2) studies with comparisons of CTO-PCI in patients with and without prior CABG; (3) studies that reported the angiographic characteristics and in-hospital complications in both groups. And case reports, reviews, notes, letters, commentaries, and editorials were excluded.

| Endpoints
The angiographic characteristics of the enrolled studies included blunt stump, degree of calcification, bending, lesion length, previously failed attempt, proximal cap ambiguity, situation of interventional collaterals, side branch at proximal entry, in-stent CTO, inadequate distal landing zone, distal cap at bifurcation, and J-CTO score. The inhospital clinical complication included perforation, tamponade, major bleeding, and contrast-induced nephropathy.

| Statistical analysis
For dichotomous data, the random-effects model with the Mantel-Haenszel method was used as a summary statistic to calculate the pooled odds ratio (OR) with the 95% confidence intervals (95% CIs).
For continuous data, standard mean differences (SMD) calculated according to the inverse-variance method was used. Statistical heterogeneity was assessed using Cochrane's Q via the chi-square test and further quantified with the I 2 statistic. And 25%, 50%, and 75% indicated low, moderate, and high heterogeneity, respectively. All statistical analyses were conducted using Review Manager 5.4.1 version (RevMan, The Cochrane Collaboration). As less than 10 studies were included in our meta-analysis, funnel plots were not used to assess publication bias. Figure 1 shows the flowchart of the study selection. As a result, a total of 2838 studies were identified through electronic searches.

| Characteristics of the studies and patients included
Then, 2829 studies were excluded after reading the titles and abstracts. The remaining nine studies were evaluated by reading the full texts. Eventually, eight studies comprising a total of 13 439 patients met the inclusion criteria and were included in qualitative synthesis and meta-analysis. [8][9][10][11][12][13][14][15] The baseline clinical characteristics of the patients included in this meta-analysis are summarized in Table 1A    The CTO lesions ratio distributions are shown in  Figure 2A).

| Proximal cap ambiguity
Proximal cap ambiguity in angiographic makes PCI more complex.

| Moderate or severe calcifications
A total of six trials reported the degree of calcification in both groups, which involved 10 905 patients. Figure 2    p < .00001), and no heterogeneity was found for bending incidence (I 2 = 0.0%, p = .8; Figure 2E).

| Lesion length
Lesion length >20 mm has been well recognized as an unfavorable  Figure 2F).

| In-stent CTO
In-stent CTO was reported in five studies from 8903 patients and 1242 (overall rate 14.0%) was found. However, no significant difference was found between the two groups: per treatment, the event rate was 14.3% (304 of 2120) in the prior CABG group and 13.8% (938 of 6783) in the non-CABG group. The pooled OR value was 0.99 (95% CI: 0.86-1.14, p = .88; Figure 2G), and there was no heterogeneity (I 2 = 0.0%, p = .44).

| Lack of interventional collaterals
The outcome occurred in 2013 events among 5208 participants   Figure 2J).

| DISCUSSION
Over the past years, CTO-PCI has received much attention as one of the major frontiers of interventional cardiology, and advisory documents underlining technical and organizational aspects have been published.
Management of CTO by PCI or CABG or medical therapy has been controversial for a very long time. 16 The summary statistic shows that of these patients, 11% undergo PCI, 40% undergo CABG, and 49% had medication. In patients without CTO, 36% underwent PCI, 28% underwent CABG, and 35% had medication. 17 Some experts prefer that CABG should be prioritized to patients with CTO because patients with CTO often have high SYNTAX scoring and also have a high probability of combining multiple  has included LCX-CTO as an independent predictor of technical failure. 23 Although, a more recent multicenter study reported similar success rates in all arteries, the patients with successful recanalization of LCX-CTO showed a higher cardiac long-term mortality compared to other arteries. 24 The lower success rate for LCX-CTO is likely associated with the frequent tortuosity and less interventional collaterals of this vessel.
Knowledge of potential caveats in treating CTO might cause patient selection; however, the decision of revascularization is done based on clinical need, not anatomical characteristics. Therefore, we further analyzed the lesion morphology of CTO target vessel.
The CTO lesion begins with the proximal cap. The proximal cap morphology, clarity, and whether there is a side branch at proximal entry decided the wire to probe the occlusion. 25 The CTO stump has two types which are tapered tip stump and blunt stump. While there are two types of pathological vascular channels that extend the CTO occluded segment which are endothelialized microchannel and microcapillary. The endothelialized microchannel, termed histologically recanalized segment, is a 160-230 µm diameter neovascularization that connects the occluded segment from the proximal to distal cap.
The microcapillary, termed nonrecanalized segment, is a <100 µm diameter capillary that passes into the vasa vasorum or small side branch which cannot span the occluded segment from the proximal to distal cap. The tapered tip stump CTO is more likely associated with a histologically recanalized segment and less likely to have a major side branch. In contrast, the blunt stump CTO is more likely to have a non-recanalized microcapillary, which means it will be more difficult to open the CTO lesion. 26 In our meta-analysis, patients with previous CABG undergoing CTO-PCI have a more blunt stump and proximal cap ambiguity, but the side branch at proximal entry has no significant difference between the two groups. The association between CABG and accelerated atherosclerosis progression was clearly elucidated in many studies. Especially in the proximal vessel, the competitive flow generated by the graft vascular may play a dominant role. Moreover, it has been hypothesized that the exposure of the distal cap to arterial pressure from the graft may even promote adverse remodeling and blunt rather than tapered morphology.
Once the wire crossed the proximal cap, it will enter the body of the CTO occluded segment. In each body of the CTO lesion, its difficulty and complexity depending on the degree of calcification, tortuosity, and length. In particular, the length of the occluded segment is the most significant factor to decide the success of crossing a CTO. The longitudinal continuity of microchannels spans around 85% of the total CTO length. 27 These histological characteristics provide the basis for multiple CTO-PCI algorithms. Some experienced operators believe that length greater than 20 mm was a major predictor of procedural failure to cross the occlusion segment than calcification, tortuosity, and blunt stump. In our meta-analysis, prior CABG is associated with more severe calcification, more tortuosity, and longer length of CTO, due either to the shrinkage or distortion of the oc- CTO, these small arteriole collaterals undergo remodeling to become muscular arteries known as arteriogenesis. 28 For appropriate interventional collaterals, the collateral channels can be septal arteries, epicardial connections but also can be graft vascular (either arterial or venous). 29 By using the backdoor to pass the guidewire from the donor artery via a collateral channel to penetrate the distal cap, the success rate of CTO-PCI has increased to >90% in the hands of CTO masters with a proper retrograde approach. 30 In our meta-analysis, the number of interventional collaterals has no significant difference between the two groups. Whereas on the other hand, bypass grafts can be used as collateral channels (even when occluded) to facilitate interventional devices via the retrograde approach. Data showed the most commonly used collaterals for the retrograde approach in prior CABG patients were septal collaterals (43%), followed by epicardial collaterals (34%), saphenous vein graft (30%), and left internal mammary artery grafts (3%). 13 Inadequate distal landing zone was defined as a distal vessel segment with a diameter less than 2.0 mm or with a diffuse lesion. 13 An inadequate landing zone represents an unfavorable feature for antegrade true-to-true lumen approach and dissection/re-entry techniques. However, bifurcation at the distal cap represents a favorable feature for retrograde approach. In our meta-analysis, for inadequate distal landing zone, prior CABG is related to significantly higher incidences, and for a distal cap at bifurcation, to our pleasant surprise, the incidence was significantly higher in patients with prior CABG which means they will have a good chance for retrograde approach. 31 These characteristics of CTO target vessel are also taken into account by risk scores for predicting technical success rate, such as the J-CTO (multicenter CTO registry in Japan) score. 32  These issues might confound the relationship between success rates and procedural complications. The method of meta-regression may solve this question. However, the meta-regression requires the number of studies included in meta-analysis not too small, and more than 10 are recommended. Otherwise, the selected influencing factors are extremely unstable. Finally, publication bias is possible that more experienced centers and higher volume might be more likely to publish their outcomes.

| CONCLUSIONS
In the previous study, successful PCI for CTO was related to increasing living quality in both patients with and without CABG history. Furtherly, interaction analyses indicated a similar revascularization profit in both groups. These results extend our thinking about benefits related to successful CTO revascularization for the high-risk patient subgroup with prior CABG and highlight the value to provide PCI for them. According to our results, CTO-PCI is frequently performed in patients with prior CABG. Nevertheless, CTO-PCI for patients who have undergone CABG is still a complex intervention and is related to higher angiographic challenges such as blunt stump, proximal cap ambiguity, severe calcifications, bending and lesion length >20 mm.
Even in the PROGRESS scores, the CABG history represents an independent risk of difficulty. Taking those specificities into account, some experts encourage that the retrograde approach should be accepted more widely, and performed appropriately. 31 And to our pleasant surprise, the rate of complications in the prior CABG group remains comparable to the non-CABG group, which makes interventionalists motivated to update their techniques and equipment.

ACKNOWLEDGMENTS
This study was supported by National Natural Science Fund of China

CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.

AUTHOR CONTRIBUTIONS
Yuchen Shi, Songyuan He, and Jinghua Liu conceived the study and designed the protocol. Yuchen Shi and Songyuan He integrated the data and drafted the manuscript. Wen Jiana and Xueqian Shen were responsible for the study selection, data extraction, and assessment of study quality. Jesse Luo and Jinghua Liu revised the manuscript critically. All authors read and approved the final manuscript.

DATA AVAILABILITY STATEMENT
Data available on request from the authors. The data that support the findings of this study are available from the corresponding author upon reasonable request.