Author + information
- Stéphane Rinfret, MD, SM∗ ( and )
- Luiz Fernando Ybarra, MD, MBA
- ↵∗Address for correspondence:
Dr. Stéphane Rinfret, McGill University Health Centre, Royal Victoria Glen Site (B03.7200), 1001 Boulevard Décarie, Montreal, Québec H4A 3J1, Canada.
- antegrade dissection and re-entry
- antegrade wire escalation
- chronic total occlusion
- percutaneous coronary intervention
- randomized control trial
The CrossBoss-Stingray antegrade dissection and re-entry (ADR) system (Boston Scientific, Natick, Massachusetts) was designed to help overcome a common failure mode of antegrade chronic total occlusion (CTO) percutaneous coronary intervention (PCI): subintimal tracking. The first pivotal study using CrossBoss-Stingray devices confirmed the safety and feasibility of ADR with these devices in selected patients, but was never compared with an antegrade wire escalation (AWE) strategy (1). One-third of CTOs could also be crossed true to true lumen without the need for the Stingray system. ADR was subsequently incorporated as part of a hybrid strategy, and is still considered a necessary and desirable approach for long CTOs with good distal target, especially when no retrograde collateral channel can be used (2). The hybrid approach has been followed effectively and safely by many operators around the world, with use of ADR in 15% to 25% of cases (3–5).
By the hybrid algorithm, AWE should be selected first for CTOs without proximal cap ambiguity and shorter than 20 mm, although longer CTOs can nowadays be crossed with new high-torque wires. In this issue of JACC: Cardiovascular Interventions, Karacsonyi et al. (6) report the results of a prospective multicenter randomized controlled trial (RCT) comparing procedural outcomes of patients undergoing CTO PCI with a planned antegrade approach, using either guidewires or the CrossBoss device. Investigators screened 966 patients from 2015 to 2017 in 11 centers and randomized 246 into a CrossBoss-first versus guidewires-first strategy. The primary efficacy endpoint of crossing time was not significantly shortened with CrossBoss (56 vs. 66 min). The primary safety endpoint of major adverse cardiac events did not differ (3.28% vs. 4.03%) between groups, and other procedural characteristics were similar. Success rates were high in moderately difficult lesions (Japan Chronic Total Occlusion score = 2).
The investigators should be commended for conducting one of the rare RCTs in CTO PCI, and to our knowledge, the first to compare 2 accepted strategies. Recruitment of patients with CTOs eligible for AWE resulted in the randomization of only one-fourth of screened patients, who had more favorable angiographic criteria: high in-stent occlusion rate, low rate of proximal cap ambiguity and good landing zone. Moreover, the use of an AWE in 22% through the CrossBoss catheter rather than using its spinning properties is another limitation and must be considered a high and difficult-to-handle crossover rate.
Most importantly, given the absence of difference in crossing time and clinical outcomes, the much higher rate of ADR techniques in the CrossBoss arm (50% vs. 22%) required to succeed has possible unmeasured clinical implications. As stated, it is perceived that the CrossBoss catheter limits the size of the dissection. We agree that the CrossBoss helps to limit the dissection plane to a smaller subadventitial track thickness or width compared with the track created by a knuckled wire or contrast. However, the CrossBoss catheter is certainly not a device designed to limit the length of the dissection. Keeping the length of the dissection plane as short as possible is desirable. Extending the length may result in significant side branch occlusion, leading to myocardial infarction (MI) and worse flow. Although the authors did not find significant difference in side branch loss, they did not report the rate of all periprocedural MI; some smaller branch occlusions could have been missed in the absence of a core lab analysis. Moreover, the CrossBoss strategy was associated with a 2.46% rate of Q-wave MI, compared with 0.81% in the AWE group; had the study been better powered for safety, such a difference could have been statistically significant. Although stent length was similar in both groups, the length of subintimal stenting per se is not reported. Song et al. (7) have showed that subintimal tracking had more evidence of dye staining or extravasation, more branch occlusion, and was associated with increased in-hospital adverse events (1.9% vs. 7.9%; p = 0.04), mostly periprocedural MI, compared with intraplaque crossing.
By using the CrossBoss early, the operator expects to perform re-entry techniques in most cases. In this study, CrossBoss catheter crossed from true-to-true lumen in only one-quarter of simpler cases, a rate lower than reported previously (1). Therefore, by using the CrossBoss with its propensity to track the subintimal space, operators give up one potentially easier option, intraplaque crossing with wires, a strategy that is successful in more than 50% of patients. Re-entry, when required, involved the Stingray in 95%, with high success, even in the AWE group, confirming it can still be done without a CrossBoss first. When the Stingray device can be used to preserve side branches, long-term outcomes with ADR are excellent (8). However, use of the Stingray was not always successful, and few subintimal tracking and re-entry techniques were required. Such last-resort re-entry techniques are unfortunately associated with worse outcomes (8). Therefore, minimizing the need for ADR in simpler cases is reasonable to avoid using last-resort re-entry maneuvers, required to bailout Stingray failure. In the absence of longer-term outcomes from this RCT, one needs to remain careful with liberal use of ADR, and the common sense of using the least traumatic techniques should prevail.
Although readers may be reassured by the short-term safety profile of a CrossBoss-first strategy, we need to reinforce the fact that the study was not powered for any hypothesis concerning safety. Given an incidence of major adverse cardiac events of 3% to 4%, a reasonable noninferiority margin would have yielded a much larger sample than the one recruited. The trend seen in the reduction of perforation and pericardiocentesis is an intriguing one at first glance: whereas 4 patients in the AWE group needed pericardiocentesis, none was required in the CrossBoss group. When reading the circumstances that led for pericardiocentesis, it is difficult to conclude of a protective effect from the CrossBoss, as many perforations resulted after a retrograde approach, balloon inflation or perforation after stenting; they were not related to wire manipulation. Intraplaque crossing with wires can sometimes be associated with more difficult subsequent plaque modification maneuvers (rotational atherectomy, high pressure inflation, cutting balloon) compared with when operators “get around” the CTO plaque with ADR. Such findings should be confirmed with a larger sample. Finally, unit costs are not provided. Given the absence of difference in aggregated costs, unit costs were likely substantially cheaper than in other countries, including in Canada where CrossBoss and Stingray add at least U.S. $2,500 in procedural costs.
The study results comfort our beliefs: wires can be successfully used to cross the CTO in >50%, do not increase procedure time nor complication, and re-entry technique can still be performed in case of subintimal tracking, even without the CrossBoss catheter. In case of failed intraplaque tracking, dissection planes can be created with polymer-jacketed knuckled wires supported by modern microcatheters. Knuckled wires more predictably track the main vessel subintimal space and stay away from branches, compared with the CrossBoss that is often advanced with much less control. The CrossBoss catheter, given its stiffness, often tracks side branches arising from the greater curvature of the coronary artery. Such branches may be smaller than the CrossBoss tip, resulting in perforation. The low perforation rate reported with the CrossBoss catheter in this study might be attributable to shorter CTOs compared with those usually tackled with ADR. After successfully tracking a subintimal plane with a knuckled wire, the microcatheter can be advanced proximal to a re-entry spot, and subsequent wires advanced straight into the subintimal space next to the true lumen. Such dissection planes with straight wires are unlikely to damage the re-entry spot. Following this, the microcatheter is advanced to dilate the track, and is subsequently exchanged for the Stingray balloon. Such AWE ± ADR as described in the preceding text, with more effective wires, has substantially decreased our use of the CrossBoss catheter to few units over the last 2 years.
CrossBoss never? We think operators still need this catheter for specific tacks. Its ability to track the true lumen for in-stent occlusions is attractive and was associated with a shorter crossing time in this subgroup compared with AWE (41 min vs. 66 min; p = 0.047). Also, although the CrossBoss is perceived to be a “poor starter” of a dissection plane, it is invariably a “good finisher” when used on a very short final distance, resulting in a relatively thin and “dry” spot lying in the outer curvature of the artery, easing subsequent re-entry attempts.
But CrossBoss first? Maybe for in-stent CTOs, although a time difference in crossing may not justify the incremental cost. In general, “CrossBoss last” in selected patients remains the most reasonable approach, and “Stingray always” when ADR is attempted.
↵∗ Editorials published in JACC: Cardiovascular Interventions reflect the views of the authors and do not necessarily represent the views of JACC: Cardiovascular Interventions or the American College of Cardiology.
Dr. Rinfret has been a consultant to Boston Scientific, Teleflex, Abbott Vascular, Abiomed, and SoundBite Medical. Dr. Ybarra has reported that he has no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Whitlow P.L.,
- Burke M.N.,
- Lombardi W.L.,
- et al.
- Brilakis E.S.,
- Grantham J.A.,
- Rinfret S.,
- et al.
- Sapontis J.,
- Salisbury A.C.,
- Yeh R.W.,
- et al.
- Maeremans J.,
- Walsh S.,
- Knaapen P.,
- et al.
- Dautov R.,
- Ribeiro H.B.,
- Abdul-Jawad Altisent O.,
- et al.
- Karacsonyi J.,
- Tajti P.,
- Rangan B.V.,
- et al.
- Song L.,
- Maehara A.,
- Finn M.T.,
- et al.
- Azzalini L.,
- Dautov R.,
- Brilakis E.S.,
- et al.