Percutaneous Coronary Intervention of Chronic Total Occlusion in Patients With Prior Coronary Artery Bypass Graft: The Current Situation

Objectives. The equipment and strategies used for percutaneous coronary intervention (PCI) for chronic total occlusion (CTO) have been improved. However, CTO-PCI for patients with prior coronary artery bypass graft (CABG) remains challenging. This study aimed to compare the strategies and initial success rates of CTO-PCI in patients with and without prior CABG.

Methods. The authors extracted data from the Japanese CTO-PCI expert registry for this study. They enrolled 11 605 patients who underwent CTO-PCI by Japanese operators from 2014 to 2022. The cohort was divided into 2 groups: post-CABG (n = 830) and no-CABG patients (n = 10775).

Results. The post-CABG patients were older than the no-CABG patients (70.3 ± 9.2 vs 67.5 ± 11.1 years, P < .01). The post-CABG group exhibited more prevalent long, tortuous, and calcified lesions, as well as higher Japanese Multicenter CTO Registry scores than the no-CABG group (1.8 ± 1.1 vs 1.4 ± 1.1, P < .01). The post-CABG patients opted for the primary and rescue retrograde approaches more frequently than the no-CABG patients (52.4% vs 40.7%, P < .02), and the post-CABG patients exhibited a lower success rate than the no-CABG patients (82.2% vs 90.2%, P < .01). However, an improvement in success rates was observed in the post-CABG patients compared with that of Japanese data from 1999 to 2011 (71%-82.2%). Additionally, the procedure time decreased from 210 to 191 minutes.

Conclusions. Compared with no-CABG patients, the initial success rate of CTO-PCI for post-CABG patients remains low, and the retrograde approach is more commonly chosen. However, the success rate has improved over previous data.

Percutaneous coronary intervention (PCI) is a valid treatment option for chronic total occlusion (CTO). Several treatment algorithms demonstrate the variability in strategies for each CTO-PCI situation, whereas the approach to non-CTO-PCI remains consistent. Additionally, optimal strategies for each situation have been recommended.^1-3 Advancements in technical approaches and equipment have enabled the use of any strategy (such as the retrograde approach or antegrade dissection and reentry) in any patient, resulting in an improved success rate. These favorable changes are due to improvements in guidewires and microcatheters. Furthermore, the practical use of computed tomography coronary angiography and intravascular ultrasound (IVUS) have also improved CTO-PCI.

Coronary artery bypass graft (CABG), a proven treatment strategy for complex coronary lesions, especially in CTO, predates the history of PCI. However, the idea of venous graft deterioration should be anticipated in our daily practice because angina recurrence may be encountered in patients with graft failure. In this situation, opting for PCI becomes a valid strategy rather than repeating CABG. However, conducting CTO-PCI for post-CABG patients remains more challenging than for patients without a history of CABG.^4,5 Data on CTO-PCI for post-CABG patients were first reported in Japan about a decade ago.⁶ However, advancements in technology and devices have continually brought many changes to CTO-PCI, leading to ongoing improvements in success rates. This study aimed to identify the current outcomes and situations of CTO-PCI for post-CABG patients and compare them with those of previous data.

The Japanese CTO-PCI expert registry is a prospective, non-randomized study that enrolled consecutive patients who underwent CTO-PCI performed by Japanese CTO operators.^2,7 The design and enrollment of the registry have been previously reported in detail.⁷ The secretariat of the registry (Clinical Research Center, Kurashiki Central Hospital, Ohara Healthcare Foundation, Okayama, Japan) manages the registry data. Data on diagnostic coronary angiograms, pre-intervention computed tomography images of coronary arteries, coronary angiograms, and IVUS images obtained during interventions was extracted. Subsequently, the data was to an independent core laboratory (the Cardiovascular Imaging Center, Aichi, Japan) for further analysis.

The authors are members of the Japanese Board of CTO Interventional Specialists, which was established in July 2013 and is composed of 46 highly experienced specialists. They recruited patients enrolled in the Japanese CTO-PCI expert registry between January 2014 and December 2022 for this sub-analysis of the registry. Clinical follow-up for these patients will continue until December 2027.

The ethics committee of all participating facilities in the Japanese CTO-PCI expert registry and the review board at Shinshu University approved this study. Written informed consent was obtained from all participants. All procedures performed in studies involving human participants adhered to the 1964 Helsinki Declaration and its subsequent amendments or comparable ethical standards.

In this study, all PCI was performed for the native vessels; PCI for the bypass graft was excluded. The definitions of CTO and angiographic analysis have been previously described.⁷ The difficulty of target CTO lesions was assessed using the Japanese Multicenter CTO Registry (J-CTO) score.⁸ Interventional collateral vessels were those collateral vessels that the operator considered to be crossable using a guidewire and microcatheter. The CTO-PCI procedures were divided into 3 groups based on the approach taken during the intervention: antegrade alone, primary bidirectional, and rescue retrograde. The primary bidirectional and rescue retrograde approaches involved attempting to cross a collateral channel with a guidewire to reach the distal end of the CTO vessel. The rescue retrograde approach involved switching from an initial antegrade attempt to cross the CTO lesion to a retrograde approach. The selection of the CTO-PCI strategy was at the discretion of the operator.

Contrast-induced nephropathy (CIN) was defined as an increase of greater than or equal to 25% and/or greater than or equal to 0.5 mg/dL in serum creatinine level compared with the baseline within 72 hours of the procedure. Technical success was defined as achieving successful guidewire passage through the CTO, resulting in a less than 30% residual diameter stenosis without major side-branch occlusion and Thrombolysis in Myocardial Infarction flow-3. Patient success was defined as achieving technical success without any evidence of in-hospital major adverse cardiac or cerebrovascular events (MACE), such as death, myocardial infarction, stroke, or revascularization, during the same admission.

In-hospital MACE, procedural time (defined as the time from initial insertion of the guidewire into the coronary lumen to the final angiography of the CTO lesion), guidewire manipulation time (defined as the time required until crossing the CTO was achieved or the procedure was aborted), and total contrast volume are recorded in the registry, and were recorded for this study.

Categorical variables are expressed as numbers (%) and compared using the chi-square test. Continuous variables are presented as mean ± SD or median (interquartile range) and were compared using the Student’s t-test or nonparametric methods. The potential predictors of CIN were evaluated and expressed them as odds ratios (ORs) and 95% CIs. Two logistic regression models were developed to evaluate CIN incidence using traditional risk factors for CIN, expressed as OR and 95% CI. All statistical analyses were performed using IBM SPSS Statistics version 26 (IBM Corporation). All reported P-values are 2-sided, with P-values of less than 0.05 considered statistically significant.

During the observation period, 11 674 patients who underwent CTO-PCI were enrolled in the CTO-PCI expert registry. Patients without a history of CABG (n = 65) were excluded, as well as 4 for whom PCI was performed for bypass grafts. After exclusions, 11 605 patients were enrolled, with 830 (7.1%) having a history of CABG. Table 1 presents the baseline characteristics of post-CABG and no-CABG participants. The post-CABG group had a significantly higher mean age than the no-CABG group (70.3 ± 9.2 vs 67.5 ± 11.1 years old, P < .01). Furthermore, the presence of hypertension, dyslipidemia, history of smoking, and diabetes mellitus were significantly associated with a history of CABG. The post-CABG group exhibited unfavorable estimated glomerular filtration rates and left ventricular ejection fractions.

The post-CABG group demonstrated a trend toward more frequent observation of the right coronary arteries (58.4% vs 49.1%, P < .01). Evaluating J-CTO score components revealed that the post-CABG group exhibited significantly more longer lesions (≥ 20 mm), tortuous lesions, and calcification, as well as higher total J-CTO scores (1.8 ± 1.1 vs 1.4 ± 1.1, P < .01).

Table 2 presents the comparison of procedural characteristics. The femoral approach was more frequently adopted in the post-CABG group. Compared with the no-CABG group, the post-CABG group exhibited longer procedure and fluoroscopic times, and significantly lower contrast media volume. The post-CABG group exhibited a lower initial patient success rate (82.2% vs. 90.2%, P < .01), as well as a higher complication rate (including death, myocardial infarction, cerebral complications, vessel perforation related complication, access site complication, and CIN) (13.2% vs 8.9%, P < .02).

The adopted procedural strategies between the post-CABG (n = 830) and no-CABG groups (n = 10 775) were compared (Figure 1). The primary bidirectional approach showed a trend toward being more frequently adopted in the post-CABG group than in the no-CABG group (36.1% vs 25.4%), whereas the strategy of the antegrade approach alone was more often observed in the no-CABG group (47.4% vs 59.3%).

Table 3 shows the difference of lesion characteristics between the lesion-with-graft or lesion-without-graft groups. Re-try cases were more frequent in lesion-without-graft group (15.7% vs 21.6%: P = .03). However, the J-CTO score was not different between the 2 groups (1.83 ± 1.17 vs 1.89 ± 1.09: P = .51). CTO-PCI strategies were different regardless of whether the target vessel was sutured graft or not. In thelesion-without-graft group, the antegrade-alone approach was used in most cases. On the other hand, the retrograde approach was more frequently employed in the lesion-with-graft group (antegrade alone: 40.2% vs 60.6%, primary bidirectional approach: 42.6% vs 24.4%, rescue retrograde approach: 17.2% vs 15.0%; P < .01). (Figure 2) The technical success rate and complication rate were not different between 2 groups (technical success rate: 84.4% vs 84.7%, P = .92; complication rate: 13.2% vs 13.2%, P = .88). However, the procedural time was longer in the lesion-with-graft group than that in the lesion-without-graft group (203.3 ± 106.1 vs 168.7 ± 98.6 minutes; P < .01). Contrast media volume was not different between groups (189.2 ± 96.1 vs 190.0 ± 91.9 mL: P = .91) (Figure 3).

Figure 4 illustrates the comparison between the previous Japanese data extracted from 1999 to 2011⁶ and our data (2014-2022). The patient success rate increased from 71% to 82.2%, and the procedure time was shortened from 210 to 191 minutes. Conversely, the retrograde approach usage was similar, at 47% and 52%.

Figure 5 compares the annual changes of technical success rate between the post-CABG group and the no-CABG group. Over all years, the technical success rate was higher in the no-CABG group than in the post-CABG group. Furthermore, in the no-CABG group, the technical success rate improved significantly from 2020 to 2023 compared with that from 2014 to 2019 (92.9% vs 90.9%, P < .01), while that of the post-CABG group was not different (87.0% vs 83.6%, P = .27).

Our study revealed that CTO-PCI for post-CABG patients was performed in 7.1% of the CTO-PCI registry population. The post-CABG group had higher J-CTO scores and more complex backgrounds. The technical and patient success rates in this group were 84.5% and 82.2%, respectively, both of which were worse than those reported in the no-CABG group. In the post-CABG group, the primary bidirectional strategy was more frequently adopted. The intervention strategy was different depending on whether the lesion did or did not have a graft. The antegrade approach was adopted more frequently in the lesion-without-graft group, and the retrograde approach was more likely to be used in the lesion-with-graft group. The J-CTO score and technical success rate were not different between the 2 groups. However, the procedural time was longer in the lesion-with-graft group. The patient success rate was improved from the 1999 to 2011 Japanese data to our 2014 to 2022 data. However, when comparing the last 9 years, even though the technical success rate of the no-CABG group was improved, that of the post-CABG group did not change.

Generally, CABG can influence the procedural difficulty of PCI because the graft suture may dislodge the native coronary artery and create severe bends. Furthermore, after CABG, pericardial tissue adhesion can cause vessel contracture, preventing vessel stretching during wire manipulation. Even though there are no differences in J-CTO scores between the post-CABG and no-CABG patients, there are specific difficulties in post-CABG cases. One difficulty is that he characteristics of the saphenous vein graft (SVG) deteriorate over time. One month after CABG, SVG disease development begins with neointimal hyperplasia, resulting in clinically significant stenosis and occlusion in the chronic phase.⁹ Previous studies have reported that 12% of SVGs will be occluded at 1 year, and 40% will be diseased within 10 years.¹⁰ Furthermore, PCI for SVG lesions has a high risk of distal embolization and restenosis because the degenerated SVGs are often enlarged and contain massive soft material-like thrombi.¹¹ In such procedures, even if vasodilators are administered or filter devices are used to prevent distal embolism, completely preventing peri-procedural infarction is challenging. Thus, PCI for native coronary arteries with diseased SVGs should be considered rather than PCI for SVGs.

Previously, Teramoto et al reported on CTO-PCI for post-CABG patients using data from 1999 to 2011. The study revealed an initial success rate of 71% in post-CABG patients and 83% in the no-post CABG patients. Our study revealed that the post-CABG group had more frequent retrograde approach usage than the no-CABG group.⁶ Several reports have compared the outcomes between post-CABG and no-CABG patients and indicated that post-CABG groups exhibited an initial success rate of 5% to 10% below that of no-CABG groups, and that retrograde approach usage was 1.5- to 2.0-fold higher than that of no-CABG groups^12-18(Table 4). The retrograde approach using a bypass graft is a special technique for post-CABG cases. In this population, 94 cases (22.1%) were performed by the retrograde approach using bypass grafts as the retrograde conduit in all of retrograde approach cases (n = 425). The difficulty of this approach depends on the condition of the graft and the angle of the sutured point. These results reflect the higher procedural difficulty in post-CABG patients. The CTO lesions in post-CABG patients are complex because CABG grafts are related to the development of atherosclerosis in the native coronary artery owing to the flow competition between the native coronary artery and the grafts.¹⁹ Additionally, our data revealed the same tendency: that long, moderately to severely tortuous, calcified lesions were more frequently observed in post-CABG patients, resulting in a higher J-CTO score than that of the no-CABG patients.

The lesions of post-CABG patients were divided into 2 groups: lesion with graft and lesion without graft. As mentioned previously, it is conceivable that graft sutures cause to the vessel. However, our data revealed there were no differences in lesion characteristics and J-CTO scores between groups. On the other hand, the interventional strategies were different: the antegrade approach was mostly commonly used in the lesion-without-graft group and the retrograde approach was more common in the lesion-with-graft group. Because CABG complicates the coronary anatomy not only of the target vessels but also the collateral donor arteries, the CTO-PCI strategy changes depending on each situation. However, the adoption of either strategy did not affect the procedural outcome.

Over the past 20 years, the initial outcomes of CTO-PCI for post-CABG patients have improved dramatically and the procedure times have been shortened. In the last decade, research has shown an 80% improvement in success rate (Table 4). This improved quality is due to recent advancements in methodology and equipment, and the accumulation and sharing of experience in this field have undoubtedly made CTO-PCI progress. However, the use of the retrograde approach remains at approximately 50%. Our data revealed that the outcome of post-CABG patients has not improved in the last few years, even though that of no-CABG patients has improved. This may indicate a current limitation of CTO-PCI. Approximately 30% patients in whom CTO-PCI was performed had a history of CABG. Therefore, the development of an alternative antegrade approach, such as an IVUS-based strategy, and related methodologies is mandatory.

In our daily practice, it is expected that we will continue to encounter bypass failure cases because of the predisposition of SVG deterioration, as mentioned previously. Given the severe background of patients with prior CABG, who are older with complex lesions and many comorbidities, PCI for native coronary lesions may be the ideal treatment and should take priority over repeat CABG and PCI for grafts.20

Limitations

This study has several limitations. First, the data were not derived from randomized analysis; thus, bias may have existed in the procedural strategies. Second, the occlusion period could not be identified for most of the study population, which is a common issue in CTO research. Third, the period of time between CABG and CTO-PCI performance was not clearly defined. Lastly, the detailed lesion morphology of the bypass graft was unclear, introducing potential variations in the classification of patent grafts, including complete patency or some material adhesion, narrowing of sutured angle, and narrowing at sutured points; these factors can influence the procedure.

Post-CABG patients have a complex background, and despite more frequent use of the retrograde approach, the success rate of CTO-PCI for these patients is still lower than that associated with no-CABG patients. Although the retrograde approach was more frequently adopted in the lesion-with-graft group, the procedural success rate did not differ depending on whether the target lesion did or did not have a graft Although outcomes have improved over the past 2 decades, as data from the last few years indicate, the improvement might stagnant. CTO-PCI for post-CABG patients is encountered frequently in daily practice and, therefore, we should continue to accumulate data and strive to improve the outcomes of post-CABG patients to equal those of patients without a history of CABG.

Soichiro Ebisawa, MD, PhD¹; Etsuo Tsuchikane, MD, PhD²; Koichi Kishi, MD³; Yoshiaki Ito, MD, PhD⁴; Hisayuki Okada, MD, PhD⁵; Satoru Sumitsuji, MD⁶; Yuji Oikawa, MD⁷; Ryohei Yoshikawa, MD⁸; Hiroyuki Tanaka, MD⁹

From the ¹Department of Cardiovascular Medicine, Shinshu University School of Medicine, Matsumoto, Japan; ²Department of Cardiology, Toyohashi Heart Center, Aichi, Japan; ³Department of Cardiology, Tokushima Red Cross Hospital, Tokushima, Japan; ⁴Division of Cardiology, Saiseikai Yokohama-City Eastern Hospital, Kanagawa, Japan; ⁵Department of Cardiology, Seirei Hamamatsu General Hospital, Japan; ⁶Department of Cardiovascular Medicine/Future Medicine, Osaka University Graduate School of Medicine, Osaka, Japan;  ⁷The Cardiovascular Institute, Tokyo, Japan; ⁸Director of Cardiology, Sanda City Hospital, Hyogo, Japan; ⁹Department of Cardiology, Kurashiki Central Hospital, Okayama, Japan.

Acknowledgments: The authors are grateful to the members of the cardiac catheterization laboratories of the participating centers and the clinical research coordinators.

Disclosures: The authors report no financial relationships or conflicts of interest regarding the content herein.

Address for correspondence: Etsuo Tsuchikane, MD, PhD, Department of Cardiology, Toyohashi Heart Center, 21-1 Gobutori, Oyamacho, Toyohashi-shi 441-8071, Japan. Email: Tsuchikane@heart-center.or.jp