Author + information
- Received April 21, 2013
- Revision received July 25, 2013
- Accepted July 29, 2013
- Published online January 1, 2014.
- ∗Peter Munk Cardiac Centre, University Health Network, Toronto, Ontario, Canada
- †San Raffaele Scientific Institute and EMO-GVM Centro Cuore Columbus, Milan, Italy
- ↵∗Reprint requests and correspondence:
Dr. Vladimír Džavík, Interventional Cardiology Program, Peter Munk Cardiac Centre, University Health Network, 6-246 EN Toronto General Hospital, 200 Elizabeth Street, Toronto, Ontario M5G 2C4, Canada.
Objectives This study sought to evaluate the feasibility of performing contemporary bifurcation techniques with the Absorb everolimus-eluting bioresorbable vascular scaffold (Abbott Vascular, Santa Clara, California) (BVS).
Background The feasibility of using the BVS in bifurcation lesions is unknown.
Methods We performed bifurcation stenting procedures including main-vessel stenting with ballooning of the side branch through the BVS struts, T-stenting and crush and culotte procedures, in a synthetic arterial model. Low-pressure final kissing balloon (FKB) inflation was performed to complete the procedures.
Results Single-stent procedures optimally opened the side-branch ostium without deforming the main vessel BVS. T-stenting completely covered the side-branch ostium. In crush cases, we could easily re-cross the crushed BVS with the wire and balloon and achieve good results after deployment of the main-vessel BVS and FKB inflation. A 2-BVS culotte resulted in good paving of the main vessel. Disruption of 1 BVS strut was observed after FKB inflation with the 2 balloons inflated beyond the recommended limit of the BVS, as calculated by Finet's law.
Conclusions Intervention of bifurcation lesions using the Absorb BVS using modern bifurcation techniques appears feasible in a coronary bifurcation model. Provisional stenting is recommended in the majority, with sequential balloon inflations and FKB inflation only when necessary. T or T-stenting and small protrusion stenting with a metal drug-eluting stent is preferable in case of crossover. A 2-BVS, T-stent technique can be performed in a high-angle bifurcation; otherwise, crush or culotte should be considered, using metal DES in the side branch. Two-BVS crush and culotte require careful evaluation, and should only be considered in patients with large-caliber main vessels.
Development of novel techniques after the introduction of drug-eluting stents (DES) (1) markedly improved outcomes after percutaneous coronary intervention (PCI) of bifurcation lesions compared with the pre-DES era (2). Bioresorbable vascular scaffolds (BVS) represent a promising new technology that theoretically can eliminate the late and very late stent thrombosis observed after deployment of metal DES because, at some point, the physical material that could potentially provide a nidus for a stent-related thrombotic event completely disappears. The most widely studied BVS, the Absorb BVS (Abbott Vascular, Santa Clara, California), a poly-l-lactic acid with promising results out to 4 years (3), has yet to be evaluated in patients with planned treatment of bifurcation lesions (4–6) because, unlike a metal stent, this polymeric BVS can unravel when deployed beyond recommended diameter limits (7). With as many as 20% of PCI patients undergoing treatment of bifurcation lesions (2,8), the generalizability of the Absorb BVS to an all-comer PCI population remains uncertain. Our aim was to evaluate the performance of the Absorb BVS in a variety of bifurcation techniques commonly in use today in an in vitro bifurcation phantom model to gain an understanding of the utility of the Absorb BVS in coronary bifurcations in the clinical setting.
All experiments were performed in the Abbott Vascular Research and Development facility in Santa Clara, California on December 4, 2012, in a synthetic arterial model composed of a polyvinyl alcohol vessel resting in a bifurcated silicone soft plate groove with a 75° bifurcation angle as measured from the axis of the main vessel to the axis of the origin of the side branch (9). The vessel lumen diameter was 3.0 mm in the main vessel and 2.5 mm in the side branch. Although this was not a fractal model, the phantoms had elastic properties that allowed stretching of the material beyond the nominal diameter. The model was immersed in an aqueous bath heated to 37°C. Guidewires, balloons, and BVS were introduced and deployed via a 6-French vascular sheath inserted in the proximal segment of the bifurcation. Balanced middle-weight guidewires (Abbott Vascular) were used for all procedures and NC Trek balloon catheters (Abbott Vascular) were used for post-dilation.
We performed provisional stenting with a final kissing balloon (FKB) (n = 2), modified T-stenting with an FKB (n = 2), double or 2-step crush (n = 2) technique, mini-crush technique (n = 1), and culotte technique (n = 1). The details of these procedures are found in the Online Appendix.
Assessment of the techniques
All procedures were assessed by visual means, and digital photographs were taken. The single-stent procedures were assessed by scanning electron microscopy and 2-stent procedures by micro computed tomography (CT), described in the Online Appendix.
Kissing balloon inflation
The BVS struts formed a short scaffold for the side-branch ostium (Fig. 1A) and its complete opening (Fig. 1B). There was no evidence of link or ring disruption in the proximal segment of the BVS. Optimal apposition was maintained throughout the length of the BVS. Scanning electron microscopy revealed a good opening of the side-branch struts without ring or link disruption (Figs. 1C and 1D).
With the 75° bifurcation angle, the side-branch ostium was adequately covered (Fig. 2). Of note, inflation of both balloons to 10 atm resulted in structural disruption of 1 proximal ring (Fig. 3A), confirmed by micro CT (Figs. 3B to 3D).
With the double-crush procedure, BVS material accumulated on the proximal surface where the crushed BVS interfaced with the overlying main vessel BVS struts (Fig. 4A). A small area of malapposition was observed between the carina and the 2 overlying scaffolds (Fig. 4A). Micro CT revealed good paving of the main vessel (Fig. 4B) with protrusion of 1 strut into the lumen without unraveling of the BVS. The side-branch ostium was completely opened (Figs. 4C and 4D). The mini-crush procedure resulted in mild protrusion of the crushed BVS into the main vessel lumen and partial protrusion into the side-branch ostium (Fig. 5).
The culotte technique procedure resulted in a thick circumferential 2-layer BVS wall in the proximal segment of the main vessel (Fig. 6A) as well as a bulky BVS neocarina (Fig. 6B), the latter likely resulting from the initial re-wiring of the side-branch BVS being at the opposite wall of the main vessel relative to the side-branch ostium and/or re-wiring of the side-branch after deployment of the main vessel BVS too proximally, thus pushing BVS material toward the carina. The small area of malapposition observed in the double-crush cases was also observed with the culotte technique (Fig. 6A). The side-branch BVS was damaged during the micro CT process, rendering assessment of the side-branch ostium with this modality impossible (Fig. 6C). The view through the main vessel lumen revealed preserved integrity of the BVS and a well-paved main vessel lumen (Fig. 6D).
Although this was not a fractal model, because of the elastic properties of the phantom, some stretching of the proximal main vessel segment occurred, thus approaching to some extent fractal conditions, as illustrated in part A of all individual technique figures.
The main findings of our study are that commonly used bifurcation techniques can be performed using the polymeric Absorb BVS, while preserving the integrity of the struts and quality of luminal reconstruction. BVS struts overlying a side branch can be re-crossed predictably with a workhorse guidewire, and balloon dilation can be performed in a manner that effectively creates an opening into the side branch. Finally, the side-branch ostium can be crossed without difficulty with another BVS through an opened BVS cell in this in vitro model. Further confirmation is needed in more challenging models before applying this maneuver to the clinical setting.
We observed no distortion of the main-vessel BVS at the bifurcation, known to occur after balloon inflation through the struts of a metal DES (10), likely because any distortion was corrected by gentle FKB inflation. In the provisional stenting strategy, optimal apposition and correction of distortion of the main-vessel BVS could also be achieved by means of sequential inflation of the balloon in the side branch and then a balloon in the main vessel.
Although the majority of patients undergoing bifurcation stenting can be treated with a single stent deployed in the main vessel (11–13), a proportion may also require stenting of the side branch. Furthermore, although FKB inflation is not always necessary, this may only apply to bifurcations with short lesions at the side-branch ostium, and only when the side branch is not compromised by the main-vessel stent (14). Thus, treatment of the side branch is often necessary, and understanding whether this can be performed with a BVS in the main vessel is essential to extending its utility to the bifurcation PCI subset.
Two-year follow-up of the Absorb A optical coherence tomography substudy showed that although the BVS struts on the proximal side of jailed side branches were completely incorporated into the vessel wall, struts overlying the distal aspect of the side branches were completely covered by a semilunar membranous neocarina, postulated by the authors to be the result of a complex interaction between the polymeric struts, the vessel wall, and shear stress (15). The extent to which this membranous neocarina might affect flow in larger side branches with more cells is unknown. It may thus be more prudent to systematically open the struts and prevent the formation of this membrane that could cause ischemia or potentially serve as a nidus for thrombus formation (15). We have shown that this can be effectively performed by crossing the BVS struts on the distal side of the side-branch ostium and that balloon inflation can create an effective opening with a scaffold of BVS material on the proximal surface of the side-branch ostium. Crossing of the distal struts overlying the side branch is recommended, as crossing more proximally and then dilating can result in the creation of a bulky mass of scaffold material at the carina (16), which could in itself result in exaggeration of shear forces, delay in re-endothelialization, and potentially thrombosis.
When we deployed a BVS in the side branch with a 75° bifurcation angle, protrusion of the side-branch BVS into the main vessel was unnecessary. With a narrower angle bifurcation, protrusion may be needed to cover the proximal side of the side-branch ostium. This is entirely dependent on the angle between the take off of the side branch and the distal main vessel, as well as the diameter of the side branch and hence the side-branch BVS or stent. The closer this angle is to 90°, the less likely it is that protrusion is needed. At the other extreme, the more acute the angle, the greater the protrusion needs to be.
With any degree of protrusion, an FKB is necessary to ensure that the protruding side-branch device does not excessively cover the lumen of either the main vessel or the side branch. Ormiston et al. (7) previously showed that aggressive post-dilation of a BVS beyond its recommended limit results in structural disruption of the scaffold. Thus, FKB inflation performed in the usual manner, with a main-vessel balloon sized for the main vessel is highly likely to result in unraveling of the BVS rings. We demonstrated that FKB inflation within an Absorb BVS is feasible, as long as the total diameter of the inflated balloon remains below the recommended maximal diameter of the particular BVS device. Using Finet's adaptation of Murray's law (proximal main vessel = 0.678 [distal main vessel diameter + side-branch diameter]), we calculated the combined diameter of a 3.0-mm NC Trek balloon and 2.5-mm NC Trek balloon inflated to 8 atm in the proximal BVS to be just >3.5 mm, slightly exceeding the recommended limit (17,18). All but 1 of the specimens with FKB inflation performed in this manner retained their structural integrity. One specimen was found to have disruption of 1 ring. With both balloons inflated to 10 atm, resulting in a combined diameter of 3.7 mm, we observed disruption of 1 scaffold ring and distortion and thinning of 2 others. Sequential moderately high-pressure inflations (10 to 15 atm) should be performed with each of the noncompliant balloons to optimize distal-vessel BVS apposition, followed by low-pressure FKB inflation to optimize the neocarina. On the basis of our observations, we recommend that the inflation be no more than 6 atm with a 3.0-mm BVS in the main vessel, and 3.0-mm and 2.5-mm NC Trek balloons inflated simultaneously in the proximal main vessel. This recommendation is specific to this specific constellation of conditions and balloons. Other balloons are likely to have different inflation characteristics. In general, the combined diameter of the 2 balloons in the main vessel as calculated according to Finet's law, should be at least 0.1 mm less than the diameter limit of the BVS.
Indeed, it may be preferable to perform the FKB inflation with the side-branch balloon inflated almost entirely in the side branch, snuggling against main-vessel balloon to protect the carina from deformation (Fig. 7).
The proximal optimization technique is now commonly performed to optimize wire and balloon re-entry into the side branch. Care must similarly be taken to use a noncompliant balloon, to inflate the balloon slowly and to <0.5 mm above the nominal diameter of the deployed BVS.
In narrow-angle bifurcations, the protrusion of the side-branch stent or scaffold after T or T-stenting and small protrusion may be excessive. Other techniques may be preferable. In the case of severely diseased side branches, it may also be preferable to secure the side branch first. Although many have abandoned the crush techniques (19), excellent long-term results have been reported using the double-kissing crush (20) and double-crush or step-crush (21) techniques. In our current experiments, we were able to re-cross the crushed BVS with a workhorse guidewire and with a noncompliant balloon and to repeat this process after we deployed the main-vessel BVS, all with ease. A short segment of the proximal main vessel on the side of the side branch contained 3 BVS layers. The combined thickness of this layer could be up to 450 μm, slightly exceeding the combined thickness of a 3-strut layer composed of the Cypher DES (Cordis Corporation, East Bridgewater, New Jersey). It is unclear as to how this multilayer scaffold would resorb and how the underlying segment would heal over time. Using the Absorb BVS, this technique could be useful in carefully selected patients with important, severely diseased side branches and with large main vessels (3.5- to 4.0-mm diameter) in which a luminal reduction of almost 0.5 mm in a short segment may be tolerated.
Similarly, all steps of the culotte technique were accomplished with ease. We observed a circumferential double BVS layer in the proximal segment, akin to BVS overlap, that resulted in a low rate of adverse events in a small cohort of patients with Absorb Extend who underwent planned BVS overlap up to 4 struts (22). By minimizing the number of overlapping struts, a culotte technique could be performed in a similar manner.
The small area of carinal malapposition observed with the crush and culotte techniques could be exaggerated by the rounding of the carina in this model. However, it could occur clinically; thus, we recommend a higher-pressure (18 to 20 atm) inflation with a noncompliant balloon the caliber of the side-branch BVS positioned with the distal end covering the ostium of the side-branch ostium to optimize apposition before performing the low-pressure hug or FKB inflation.
Gogas et al. (23) previously reported the feasibility of re-crossing and dilating a small-caliber side branch with a 1.5-mm balloon in a patient who had ischemia and poor flow in a small side branch due to carina shift after BVS deployment. Our work suggests that this and other commonly used bifurcation techniques may be feasible in a broader spectrum of patients undergoing bifurcation PCI.
We have demonstrated the feasibility of PCI of bifurcation lesions with the Absorb BVS using modern bifurcation techniques in a phantom model of a coronary bifurcation. On the basis of these observations, our current recommendations are to use provisional stenting in the majority of cases, with sequential noncompliant balloon inflations in the side branch and then main vessel, reserving FKB inflation only for cases in which it is absolutely deemed necessary. If crossover to a 2-stent technique is necessary, T-stenting or T-stenting and small protrusion with a metal DES is preferable. If a planned 2-stent technique is deemed necessary, in a high-angle bifurcation such as in our model (≥75°), a BVS deployed in the side branch followed by a BVS in the main vessel should be feasible. In narrower angle bifurcations, a mini-step crush or culotte technique should be considered, deploying a metal DES in the side branch. Although planned bifurcation techniques such as crush and culotte using the BVS platform in both the main vessel and side branch appear feasible, their use requires careful evaluation in fractal models with different bifurcation angles and clinically should be limited to patients with large-caliber main vessels. As disrupted BVS struts cannot be visualized by angiography, we recommend intravascular imaging, preferably with optical coherence tomography, or, alternatively, with intravascular ultrasound of the main vessel, in particular, whenever dilation of the BVS struts, proximal optimization technique, or FKB inflation has been performed, or 2 BVS have been deployed, to ensure the integrity of the final result.
Dr. Džavík has received unrestricted research grants, speaker honoraria, and travel grants from Abbott Vascular. Dr. Colombo has reported that he has no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- bioresorbable vascular scaffold(s)
- computed tomography
- drug-eluting stent(s)
- final kissing balloon(s)
- optical coherence tomography
- percutaneous coronary intervention
- Received April 21, 2013.
- Revision received July 25, 2013.
- Accepted July 29, 2013.
- American College of Cardiology Foundation
- Serruys P.W.,
- Onuma Y.,
- Ormiston J.A.,
- et al.
- Diletti R.,
- Serruys P.W.,
- Farooq V.,
- et al.
- Ormiston J.A.,
- De Vroey F.,
- Serruys P.W.,
- Webster M.W.
- Colombo A.,
- Bramucci E.,
- Sacca S.,
- et al.
- Hildick-Smith D.,
- de Belder A.J.,
- Cooter N.,
- et al.
- Steigen T.K.,
- Maeng M.,
- Wiseth R.,
- et al.
- Niemela M.,
- Kervinen K.,
- Erglis A.,
- et al.
- Okamura T.,
- Onuma Y.,
- Garcia-Garcia H.M.,
- et al.
- Chen S.L.,
- Santoso T.,
- Zhang J.J.,
- et al.
- Gogas B.D.,
- van Geuns R.J.,
- Farooq V.,
- et al.