Author + information
- Received March 13, 2017
- Revision received May 9, 2017
- Accepted June 1, 2017
- Published online August 7, 2017.
- Lorenzo Azzalini, MD, PhD, MSca,
- Pierfrancesco Agostoni, MD, PhDb,∗ (, )
- Susanna Benincasa, MDa,
- Paul Knaapen, MD, PhDc,
- Stefan P. Schumacher, MDc,
- Joseph Dens, MD, PhDd,e,
- Joren Maeremans, MScd,e,
- Adriaan O. Kraaijeveld, MD, PhDf,
- Leo Timmers, MD, PhDf,
- Michael Behnes, MDg,
- Ibrahim Akin, MDg,
- Aurel Toma, MDh,
- Franz-Josef Neumann, MDh,
- Antonio Colombo, MDa,
- Mauro Carlino, MDa and
- Kambis Mashayekhi, MDh
- aInterventional Cardiology Unit, Cardio-Thoracic-Vascular Department, San Raffaele Scientific Institute, Milan, Italy
- bDepartment of Cardiology, St. Antonius Hospital, Nieuwegein, the Netherlands
- cDepartment of Cardiology, VU University Medical Center, Amsterdam, the Netherlands
- dFaculty of Medicine and Life Sciences, Universiteit Hasselt, Hasselt, Belgium
- eDepartment of Cardiology, Ziekenhuis Oost-Limburg, Genk, Belgium
- fDepartment of Cardiology, University Medical Centre Utrecht, Utrecht, the Netherlands
- gFirst Department of Medicine, University Medical Centre Mannheim, Faculty of Medicine Mannheim, University of Heidelberg, Mannheim, Germany
- hDivision of Cardiology and Angiology II, University Heart Center Freiburg Bad Krozingen, Bad Krozingen, Germany
- ↵∗Address for correspondence:
Dr. Pierfrancesco Agostoni, St. Antonius Hospital, Department of Cardiology, Koekoekslaan 1, 3435 CM Nieuwegein, the Netherlands.
Objectives The aim of this study was to describe the procedural aspects and outcomes of retrograde chronic total occlusion (CTO) percutaneous coronary intervention (PCI) through ipsilateral collateral channels (ILCs).
Background Retrograde CTO PCI via ILCs is rarely performed, usually when no other retrograde options exist, and available evidence derives mostly from case reports.
Methods A large retrospective multinational registry was compiled, including all consecutive patients undergoing retrograde CTO PCI through ILCs at 6 centers between September 2011 and October 2016. Success rates, as well as procedural complications and in-hospital outcomes, were studied.
Results A total of 126 patients (17% of all retrograde CTO PCIs) were included. The mean age was 65.7 ± 11.2 years, and the mean J-CTO (Multicenter CTO Registry in Japan) score was 2.36 ± 1.13. The target vessel was the circumflex coronary artery in 42%, the left anterior descending coronary artery in 39%, and the right coronary artery in 19%. The ILCs used were epicardial in 76% and septal in 24%. ILC anatomy was very heterogeneous. One guiding catheter was used in 80%, whereas the ping-pong technique was used in 20%. A retrograde wire could be advanced to the distal cap in 81%. Technical and procedural success rates were 87% and 82%, respectively. ILC perforation with need for intervention was observed in 5.6% and tamponade due to ILC perforation in 2.4%. One patient (0.8%) died.
Conclusions Retrograde CTO PCI through ILCs is a challenging intervention that can be performed in difficult occlusions with high success rates and reasonable rates of complications by experienced operators.
- chronic total occlusion
- collateral channels
- percutaneous coronary intervention
The retrograde approach has currently become an established, safe, and effective technique to percutaneously recanalize coronary chronic total occlusions (CTOs) (1–3) and has produced a relevant increase in procedural success rates, compared with an antegrade-only approach (from ∼80% in selected cases to ∼90% in all comers), despite a slightly increased risk for major adverse cardiac events (2,4).
Retrograde CTO percutaneous coronary intervention (PCI) is most frequently performed through septal collateral channels (1–3), as the use of this route is associated with a lower incidence of complications (periprocedural myocardial infarction [MI] and collateral channel injury), compared with epicardial collateral channels (3). For this reason, the retrograde approach through epicardial collateral channels is currently considered among the most challenging scenarios for CTO PCI operators.
In this setting, a particular subgroup is represented by retrograde CTO PCI via ipsilateral collateral channels (ILCs). ILCs are present in fewer than one-third of coronary angiographic studies of patients with CTO (5,6). Procedures via this route are associated with unique challenges, including steeper angles to be overcome with wires and microcatheters, as well as the need for re-entry into the same or an additional guiding catheter engaging the same coronary ostium (the “ping-pong” technique ). Additionally, most of these channels have an epicardial course, and as such, they are quite fragile, with a potential risk for cardiac tamponade in case of perforation. For these reasons, retrograde CTO PCI via ILCs is currently rarely performed, usually when no other retrograde options exist, and available evidence derives mostly from case reports (7–9) and the experience of 2 seasoned operators (6,10).
The aims of this study were to analyze the outcomes of retrograde CTO PCI via ILCs in a large retrospective multinational registry involving several operators and to provide useful recommendations to promote wider adoption of this advanced technique.
All patients undergoing attempted retrograde CTO PCI through ILCs at the 6 participating centers between September 2011 and October 2016 were retrospectively included. All procedures were indicated according to the presence of angina, ischemia, or both and were performed electively (ad hoc PCI was discouraged) (4) by experienced operators (>150 retrograde CTO PCIs overall or >40 retrograde CTO PCIs per year). The use of the retrograde approach in our study followed the hybrid algorithm (11). As such, it was chosen in case of: 1) proximal cap ambiguity; 2) poor distal target; or 3) presence of interventional collateral channels. Baseline, procedural, and hospitalization data were recorded. All subjects gave informed consent for the procedure, and because of the retrospective nature of the registry, the requirement for ethics committee approval was waived.
CTO was defined as a 100% stenosis with TIMI (Thrombolysis In Myocardial Infarction) flow grade 0 for >3 months (12). The J-CTO (Multicenter CTO Registry in Japan) score (13) was calculated for each lesion. Difficulty navigating the ILCs was graded according to the score proposed by McEntegart et al. (5), which assigns 1 point for each of 8 adverse collateral channel characteristics (epicardial channel, tortuous channel, small channel, long channel, adverse channel entry, adverse channel exit, multiple bifurcations, and high risk for damage or ischemia). In the original publication, mean scores for collateral channels considered to be interventional versus noninterventional were 2.18 and 4.43, respectively.
Technical success was defined as a residual stenosis <30% with antegrade TIMI flow grade 3 in the CTO target vessel (12). Procedural success was defined as technical success plus the absence of in-hospital adverse events (all-cause death, Q-wave MI, stroke, recurrent angina requiring target-vessel revascularization with PCI or coronary artery bypass graft, tamponade requiring pericardiocentesis or surgery) (12).
Procedural complications and in-hospital adverse events included procedure-related death, procedure-related stroke, periprocedural type 4a MI (14), stent thrombosis, need for urgent revascularization, major bleeding (bleeding requiring transfusion, vasopressors, surgery, or percutaneous intervention), vascular complications, myocardial ischemia due to collateral channel injury or obstruction requiring interruption of the procedure, coronary perforation requiring intervention (coil embolization, covered stent implantation, pericardiocentesis, or surgery), and contrast-induced nephropathy (increase in serum creatinine >25% or >0.5 mg/dl at 48 h post-procedure).
Continuous variables are presented as mean ± SD, and the Student t test was used for comparisons. Categorical variables are presented as frequency (percentage) and were compared using chi-square tests. Statistical analysis was performed using SPSS version 24 (IBM, Armonk, New York).
During the study period, a total of 1,670 CTO PCIs were performed at the 6 participating centers. In 721 (43%) of these, the retrograde approach was used. Among retrograde cases, ILCs were used in 126 patients (17%), who represented the study population. Table 1 shows the clinical characteristics of this cohort. The mean age was 65.7 ± 11.2 years, and 90% of patients were male. Cardiovascular risk factors and comorbidities were highly prevalent. In particular, the prevalence of diabetes was 37%, 48% of patients had experienced prior MI, and 21% had previously undergone coronary artery bypass grafting. Moderate to severe angina was present in 45%. Left ventricular ejection fraction was mostly normal (52.3 ± 12.7%). The most frequent indication for CTO PCI was angina.
Angiographic data are shown in Table 2. The most frequent target CTO vessel was the circumflex coronary artery (42%), followed by the left anterior descending coronary artery (LAD) (39%) and the right coronary artery (RCA) (19%). Importantly, a left-dominant coronary circulation was observed in 22 of 53 circumflex CTO cases (42%), ischemia in the circumflex territory was proved in 13 subjects (25%), and refractory angina despite optimal medical therapy was diagnosed in all these patients. Occlusion complexity was very pronounced (mean J-CTO score 2.36 ± 1.13). In particular, the angiographic characteristics that usually dictate the use of a retrograde approach were highly prevalent in our cohort: proximal cap ambiguity (68%), distal cap at bifurcation (37%), and distal vessel of suboptimal quality (40%).
The majority of collateral channels were classified as CC1 (69%) and CC2 (27%). Seventy-six percent of ILCs were epicardial, whereas 24% were septal. Collateral channel tortuosity was mild to moderate in the great majority of patients, with only 6% being classified as “corkscrew.” The mean McEntegart collateral channel score was 3.71 ± 1.30, which indicates considerable difficulty navigating these ILCs. The anatomic distribution of the ILCs in our study was very varied (Figure 1). The most frequent ILCs for the left coronary artery were from the circumflex system to the diagonal/LAD system or vice versa (22%), diagonal to diagonal/LAD (20%), from the apical LAD to the circumflex system or vice versa (19%), and from the LAD via septal channels to the circumflex system or vice versa (17%). All 5 types of RCA ILCs according to the classification of Mashayekhi et al. (6) were encountered and used, with type A (acute marginal to posterolateral/posterior descending artery) being the most frequent (38%).
The retrograde approach was chosen as first intention in 18 patients (14%) and as bailout following failed antegrade techniques in 108 subjects (86%). Table 3 displays the procedural data of our cohort. At least 1 femoral access was used in the majority of cases (87%). In 80% a single guiding catheter was used, whereas in 20% the ping-pong technique with 2 catheters was used. Large guiding catheters (≥7 F) were used in 90% of cases as the initial approach. When the ping-pong technique was performed, a 6-F guiding catheter was used in 7 cases (28%), a 7-F guiding in 16 (64%), and an 8-F catheter in 2 (8%). The Corsair microcatheter (64%) and the Sion family of guidewires (90%; both by Asahi Intecc, Nagoya, Japan) were most frequently used to navigate the ILCs. There were no differences with regard to wire or microcatheter use between epicardial and septal collateral channels. The operator could advance a retrograde wire to the distal cap through an ILC in 81% of cases.
Externalization was most commonly performed using the conventional technique (e.g., using an RG3 guidewire; Asahi Intecc). The most common final crossing techniques were retrograde true to true (36%), followed by reverse controlled antegrade and retrograde subintimal tracking (32%), whereas bailout antegrade techniques were successful in 11% after retrograde failure. Technical success was achieved in 87% and procedural success in 82%. There were no differences with regard to McEntegart score in cases with procedural success versus failure (3.66 ± 1.25 vs. 3.91 ± 1.54, p = 0.40). However, we observed a direct relationship between operator volume of retrograde CTO PCI through ILCs and success rates (technical success: ≤10 procedures 72% vs. 11 to 20 procedures 81% vs. >20 procedures 93%, p = 0.04; procedural success: 61% vs. 78% vs. 88%, p = 0.03). Procedural metrics reflected the high complexity of these occlusions and the retrograde approach through ILCs.
Procedural complications and in-hospital adverse events are outlined in Table 4. Tamponade was diagnosed and treated in 5 patients overall: in 3 patients (2.4%), this was due to ILC perforation, whereas in 2 cases (1.6%), it was related to target CTO vessel perforation. Major perforation requiring intervention was observed in 7 ILC patients (5.6%), and in 3 other patients (2.4%), it involved the target CTO vessel. No case of ischemia causing ST-segment elevation due to collateral channel damage or obstruction was observed during the procedure. Biochemical evidence of periprocedural MI was diagnosed in 7.1%; in all but 1 case, this was an asymptomatic event with no clinical consequences. One patient presented with symptomatic periprocedural ST-segment elevation MI and urgent need for reintervention because of acute occlusion of the LAD following retrograde extension of a dissection 1 hour after successful proximal circumflex CTO recanalization with reverse controlled antegrade and retrograde subintimal tracking through a diagonal-to-marginal ILC. The patient underwent successful additional stenting of the left main and LAD. One patient died because of presumed aortic dissection. This 27-year-old man had been diagnosed with an unspecified connective tissue disease before CTO PCI. During the index procedure, he experienced multiple coronary dissections. These dissections were observed following nontraumatic guiding catheter engagement in both the RCA and the LAD (target CTO vessel) and required the implantation of a total of 11 stents. After successful LAD recanalization through a marginal-to-diagonal ILC, the patient was admitted to the coronary care unit, where a few hours later he experienced chest pain with subsequent cardiac arrest. No echocardiographic signs of tamponade were observed. Electrocardiographic monitoring before cardiac arrest did not show ST-segment changes. Resuscitation efforts proved unsuccessful, and consent to perform an autopsy was denied by the family. Therefore, the most likely cause of death was judged to be subacute aortic dissection in the context of connective tissue disease. Other complications included contrast-induced nephropathy (n = 2), vascular complications (n = 3), and major bleeding (n = 3).
We report angiographic and procedural data as well as in-hospital outcomes of the first large multicenter registry of retrograde CTO PCI performed via ILCs. Despite complex occlusions, the operators were able to reach technical and procedural success rates in line with the figures reported by large all-comers CTO PCI registries (4), and which compared favorably with the retrograde cohorts of a European CTO registry (75% and 71%), a multicenter US registry (85% and 82%), and a Japanese multicenter registry (84%) (1–3) (Figure 2). In our experience, retrograde CTO PCI via ILCs is a challenging procedure (as reflected by high J-CTO and McEntegart scores, as well as procedural metrics) that was associated with non-negligible rates of collateral channel damage with need for intervention and tamponade. It must be underlined that 76% of ILCs in our study were epicardial channels, which are known to be at higher risk for complications than septal collateral channels, which contributes to explaining our findings. Perforation-related events were also reported, albeit to a lower extent, in the aforementioned multicenter registries on retrograde CTO PCI (perforation 2.0% to 11.7%, tamponade 0.5% to 1.3%) (1–3). However, such cohorts included retrograde procedures performed through the full spectrum of collateral channels (not limited to ILCs), and epicardial channels were used in a minority of cases (13% to 34%) (1–3). With the exception of tamponade, in-hospital outcomes of our cohort compared favorably with those from the aforementioned reports (1–3) (Figure 2).
Retrograde CTO PCI through ILCs shows distinctive characteristics. First, the prevalence of circumflex CTO is highest, and that of RCA CTO is lowest, which is opposite to vessel distribution in CTO PCI in general and overall retrograde CTO PCI. This is because the RCA and LAD most frequently receive collateral channels from the contralateral system (5), and the RCA presents very few branches proximal to the crux cordis (which could potentially give origin to ILCs). Therefore, the RCA has the lowest prevalence of ILCs, whereas the circumflex coronary artery shows the highest representation of such connections (5). Additionally, retrograde CTO PCI via ILCs is technically more challenging because of steep angles to overcome and the need for re-entry into the same or an additional guiding catheter. Finally, circumflex CTO PCI has been identified as an independent predictor of technical failure (15). As such, retrograde CTO PCI via ILCs is considered among the most challenging procedures in interventional cardiology and thus requires specific skills.
Figure 3 outlines key factors for an optimal approach to retrograde CTO PCI through ILCs. The introduction of dedicated low-profile microcatheters with improved crossing performance (16) has allowed use of the retrograde approach through epicardial collateral channels (which represent the majority of ILCs), because over-the-wire balloons are contraindicated in such setting because of their high crossing profile and stiffness, as well as poor trackability, which would expose the patient to an unacceptably high rate of collateral channel injury. In parallel, the development of novel guidewires with atraumatic tip, flexible shaft, high lubricity, and improved torque response has allowed safe tracking of very tortuous and fragile collateral channels. As a consequence, such microcatheters and guidewires are essential for operators willing to tackle retrograde CTO PCI through ILCs.
Additionally, operators willing to attempt retrograde CTO PCI through ILCs should possess advanced skills that arise from years of dedicated training, attendance at CTO meetings, and networking with peers. We advocate that such operators must be already confident performing retrograde CTO PCI through septal collateral channels and bypass grafts. Also, they must have some experience using contralateral epicardial collateral channels, because most ILCs are epicardial. We observed a direct relationship between operator volume of retrograde CTO PCI through ILCs and success rates, which had also been reported by others in overall CTO PCI (17). This emphasizes the importance of achieving and maintaining adequate skills with this complex intervention. Importantly, ILC operators must be skilled at complication troubleshooting. Perforation management includes mastering pericardiocentesis, the ping-pong technique (7), implantation of covered stents (which are challenging to deliver because of their high crossing profile), coil or fat embolization, and use of hemodynamic support devices (18). Safer navigation through collateral channels might be achieved by the use of a gentle tip injection through the microcatheter, when collateral course is not completely understood. Additionally, ischemia and subsequent hemodynamic compromise might arise when a dominant ILC is accessed. This is secondary to the shear stress induced by manipulating guidewires and microcatheters making a tight loop in a very localized region of the heart (18). However, we did not observe any of such cases in our cohort. This risk can be reduced by using alternative externalization techniques, such as tip-in (19) or rendez-vous (20) (for single-guiding-catheter procedures) or conventional externalization using the ping-pong technique (7). Also, the operator must pay attention to avoid an ostial dissection, which is expected to be more common in retrograde procedures, because of the friction created advancing the microcatheter back into the guiding catheter after retrograde crossing. Finally, we observed a non-negligible rate of periprocedural MI (7.1%), which was also reported in other cohorts treated with the retrograde approach, indicating a higher risk for myocardial injury secondary to collateral channel manipulation (21–23).
First, modeling the clinical reasoning that led the operator to choose an ILC (instead of a contralateral collateral channel) or a specific ILC is somewhat complex and was therefore not captured by our database. Second, our follow-up was limited to the hospitalization when CTO PCI through ILCs was performed. However, because all complications related to ILC use are diagnosed during the immediate periprocedural period, it is unlikely that the availability of such information would have added significantly to our study. Finally, the procedures described in this report were performed by experienced CTO operators, and our study findings might therefore not be generalizable to other clinical practices.
Retrograde CTO PCI through ILCs is a challenging intervention that can be performed with high success rates and reasonable complication rates in the hands of experienced operators. Dedicated material as well as specific continuous learning and training are fundamental to achieving such results.
WHAT IS KNOWN? ILCs are frequently encountered in CTO PCI. However, their use involves several challenges and a higher risk for complications. As such, retrograde CTO PCI through ILCs is rarely performed.
WHAT IS NEW? This multinational registry of patients undergoing retrograde CTO PCI through ILCs included 126 patients from 6 institutions. The ILCs used were epicardial in 76% and septal in 24%, and their anatomy was very heterogeneous. Technical and procedural success rates were 87% and 82%, respectively. The most frequent complication was collateral channel perforation with or without tamponade.
WHAT IS NEXT? The development of newer generation microcatheters and guidewires, as well as dedicated training initiatives for CTO PCI operators, will likely improve the outcomes of this challenging procedure. Further data are needed to ascertain the comparative effectiveness and safety of CTO PCI through ILCs versus contralateral collateral channels.
Mr. Maeremans is a researcher for the Limburg Clinical Research Program UHasselt-ZOL-Jessa, supported by the foundation Limburg Sterk Merk, Hasselt University, Ziekenhuis Oost-Limburg, and Jessa Hospital. Dr. Azzalini has received honoraria from Guerbet; and has received research support from ACIST Medical Systems. Dr. Agostoni has received honoraria from Aquilant, Meril, Neovasc, Genae, and Angiodynamics. Dr. Dens has received consulting fees from Asahi, Boston Scientific, Medical Solutions, Orbus Neich, Terumo; and has received a scientific grant from Boston Scientific. Dr. Mashayekhi has received consulting fees from Asahi Intecc; and has received honoraria from Vascular Solutions. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- chronic total occlusion
- ipsilateral collateral channel
- left anterior descending coronary artery
- myocardial infarction
- percutaneous coronary intervention
- right coronary artery
- Received March 13, 2017.
- Revision received May 9, 2017.
- Accepted June 1, 2017.
- 2017 American College of Cardiology Foundation
- Galassi A.R.,
- Sianos G.,
- Werner G.S.,
- et al.
- Karmpaliotis D.,
- Karatasakis A.,
- Alaswad K.,
- et al.
- Mashayekhi K.,
- Behnes M.,
- Akin I.,
- Kaiser T.,
- Neuser H.
- Chung T.S.,
- Agostoni P.
- Benedetto D.,
- Agostoni P.
- Mashayekhi K.,
- Behnes M.,
- Valuckiene Z.,
- et al.
- Brilakis E.S.,
- Grantham J.A.,
- Rinfret S.,
- et al.
- Morino Y.,
- Abe M.,
- Morimoto T.,
- et al.
- Christopoulos G.,
- Kandzari D.E.,
- Yeh R.W.,
- et al.
- Tsuchikane E.,
- Katoh O.,
- Kimura M.,
- Nasu K.,
- Kinoshita Y.,
- Suzuki T.
- Maeremans J.,
- Walsh S.,
- Knaapen P.,
- et al.
- Azzalini L.,
- Brilakis E.S.
- Vo M.N.,
- Ravandi A.,
- Brilakis E.S.
- Lo N.,
- Michael T.T.,
- Moin D.,
- et al.
- Lee S.W.,
- Lee P.H.,
- Kang S.H.,
- et al.