Multi-Lesion "Culotte" and "Crush" Bifurcation Stenting with Sirolimus-Eluting Stents: Long-Term Angiographic Outcome

Case Report. A 63-year-old man, an ex-smoker with a history of hypertension and previous myocardial infarction, was admitted with stable angina (Canadian Cardiovascular Society Class 1) for elective percutaneous coronary intervention. Pre-procedure coronary angiogram revealed diffuse disease in the proximal and mid segments of the left circumflex artery (LCx) that was totally occluded in its distal portion (Fig. 1A). The first obtuse marginal branch (OM) presented a severe ostial stenosis (Fig. 1A). Also, the left anterior descending artery (LAD) presented a long stenosis in its mid portion, involving the origin of the first and second diagonal branches (Fig. 2A). The right coronary artery (RCA) showed mild irregularities without any localized significant stenosis. The left coronary was cannulated with a 7 French (Fr) Vista Brite Amplatz Left guiding catheter (Johnson & Johnson, Cordis Corporation, Miami Lakes, Florida). The ostial lesion in first OM was crossed with a PT Graphix Intermediate 0.014´´ guidewire. Another PT Graphix Intermediate 0.014´´ guidewire was inserted in the LCx, but the total occlusion in its distal segment could only be partially recannalized. A 2.5 x 18 mm sirolimus-eluting Cypher stent (Cordis Corp.) was deployed (14 atm) in a stenotic lesion in the LCx just proximal to the site of the vessel occlusion in an attempt to facilitate further measures to recannalize the vessel; the stent was deployed covering the origin of the first marginal branch (Fig. 1B). A 3.0 X 18 mm sirolimus-eluting Cypher stent (Johnson & Johnson, Cordis Corp.) was then implanted in the LCx-OM using the “culotte” technique (20 atm)1 (Figs. 1C and 1D). A residual stenosis in the proximal LCx was treated with an additional 3.0 x 18 mm sirolimus-eluting Cypher stent overlapping the distal stent (20 atm). Further attempts to recannalize the distal LCx were unsuccessful. The final result is depicted in Figure 1E. Two PT Graphix Intermediate 0.014´´ guidewires were inserted in the LAD and second diagonal branch. A “crush” stent implantation was performed: a 2.25 x 8 mm sirolimus-eluting stent and a 3 x 33 mm sirolimus-eluting stent (Cypher) were concomitantly positioned in the second diagonal and mid LAD, respectively (Figure 2B). The 2.25 x 8 mm sirolimus-eluting stent was deployed (12 atm) with its proximal portion partially placed through the LAD (Figure 2B). Importantly, the LAD-diagonal stent was implanted while the undeployed 3 x 33 mm sirolimus-eluting-stent was already positioned in the LAD in the site of its future implantation (covering the origin of the diagonal branch) (Fig. 2B). Subsequently, the balloon-catheter was retrieved from the diagonal branch and the LAD stent was deployed (22 atm) (Fig. 2B). After implantation of the LAD stent, a residual stenosis was noted in the proximal stent edge together with ostial compromise of the first diagonal branch, possibly due to plaque shiftening towards its origin (Fig. 2C). Additional “crush” stenting was performed to treat the LAD-first diagonal bifurcation (Fig. 2D). Following the same strategy as describe above, a 2.25 x 8 mm sirolimus-eluting Cypher stent was implanted in the first diagonal (12 atm) with its proximal portion partially deployed in the LAD (Fig. 2D) while an undeployed 3 x 8 mm sirolimus eluting-stent was already placed in the LAD (along the diagonal ostium) (Fig. 2D). The LAD stent was then deployed (20 atm) covering the ostium of the first diagonal branch (Fig. 2D). A small gap between the two LAD stents was noted and a 3 x 8 mm sirolimus-eluting Cypher stent was implanted (20 atm) to accomplish complete lesion coverage (Fig. 2E). Final high-pressure post-dilatation (22 atm) was performed in the mid LAD (Maverick balloon 3.0 x 9 mm, Boston Scientific, Maverick, Massachusetts). Care was taken to inflate the balloon inside the stented area in order to avoid vessel injury in the non-stented edges. Excellent final angiographic result was achieved with TIMI 3 grade flow and minimal residual stenosis in all treated lesions (Fig. 2F). Figure 2 Legend: (A) Left anterior descending artery (cranial right anterior oblique projection) presenting a long stenosis in its mid portion, involving the origin of the first and second diagonal branches. (B) “Crush” stenting: a 2.25 X 8 mm sirolimus-eluting stent and a 3 X 33 mm sirolimus-eluting Cypher stent (Johnson & Johnson, Cordis Corporation) concomitantly positioned in the second diagonal and mid LAD, respectively. Note that the proximal portion of the stent placed in the diagonal is protruding into the LAD. (C) Residual stenosis noted in the proximal LAD together with ostial compromise of the first diagonal branch after implantation of the distal stent in the LAD. (D) “Crush” stenting in the LAD-first diagonal bifurcation: a 2.25 X 8 mm sirolimus-eluting Cypher stent (Johnson & Johnson, Cordis Corporation) in first diagonal branch is positioned with its proximal portion partially located though the LAD. An undeployed 3 X 8 mm sirolimus eluting-stent was concomitantly positioned in the LAD along the diagonal ostium. (E) A 3 x 8 mm SES deployed at a small gap between the two LAD SES. The radiopaque stents implanted in the 2 previously treated bifurcations are noted. (F) Final angiographic result of the treatment of the LAD with a TIMI 3 grade flow and minimal residual stenosis in all treated lesions. The patient was included in the Rapamycin Eluting-Stent Evaluated At Rotterdam Cardiology Hospitals (RESEARCH) registry and agreed to undergo late angiographic follow-up.2 After 208 days, the patient remained asymptomatic and a stress test was negative. No adverse events had occurred. At coronary angiography, the RCA was unchanged. All stents were widely patent with no angiographic evidence of stenosis (Figs. 3A and 3B). References 1. Chevalier B, Glatt B, Royer R, et al. Placement of coronary stents in bifurcation lesions by the “culotte” technique. Am J Cardiol 1998;82:943–949. 2. Lemos PA, Lee C, Degertekin M, et al. Early outcome after sirolimus-eluting stent implantation in patients with acute coronary syndromes. Insights from the Rapamycin-Eluting Stent Evaluated At Rotterdam Cardiology Hospital (RESEARCH) registry. J Am Coll Cardiol 2003;41:2093–2099. How Would you Manage this Case? Bernhard Meier, MD Swiss Cardiovascular Center Bern Bern, Switzerland The approach to this 60-year-old man with stable angina after a MI (presumably in the region of the posterolateral marginal branch of the left circumflex coronary artery) selected by the authors shows virtuosity and fantasy in the use of coronary stents. It can be condoned as exclusively drug-eluting stents (DES) with a low restenosis potential used. Also, the excellent long-term result proves the strategy right. The costs, however, are of concern and a far less expensive simple approach as described below would also have had a good chance of good long-term patency of all important branches. Working exclusively with 5 Fr coronary catheters (often inserted without an introducer thereby keeping the puncture hole smaller than that of a 4 Fr introducer), kissing balloon techniques are out of the question. The same holds true for “crush” stenting. I would have used a left Amplatz II catheter exactly like the authors, but 2 Fr smaller. I would then have first tried to recanalize the posterolateral marginal branch of the LCx. The authors failed on that, so would have I presumably. This would have reduced the bifurcation lesion in the LCx to a single vessel lesion, which I would have dilated with a 3 mm balloon and stented with a DES if the result had not been pleasing. Given the initial lesion, this would have been the case with a 60% probability. I would have used the LAD approach second, because you want a virgin coronary guidewire and balloon catheter to tackle a chronic occlusion and you also want the potential benefit of reversed collaterals from an initially recanalized LCx if problems arise in the LAD. I would have dilated the segment of the LAD encompassing the 2 involved diagonal branches. With a probability of about 30%, I would have found a need to stent the entire segment again using a DES. After this, I would have turned my attention to the 2 diagonal branches. I would have first tried to cross them with the wire and 3 mm balloon used for all dilatations so far. if they looked significantly stenosed. Only if a take-off stenosis of more than 50% of these diagonal branches had persisted, I would have tried to stent one or both take-offs with a short 2.5 mm stent, again using drug-eluting stents. The likelihood of this need can be guessed at 30% each. After such diagonal stenting, almost unvariably the 3 mm balloon would have had to be reinserted into the LAD for a final dilatation of this main axis, still using the same guidewire. In a follow-up angiogram, the chances of the LAD or the only important branch of the LCx to be restenosed would have been the same as with the technique described by the authors. The risk of a restenosis in one or both of the diagonal branches is projected at about 50%, if they had not been stented, and about 20% if they had been stented. The risk of an abrupt closure of an important vessel with the frugal approach should not have been increased (it might even have been decreased), as stents harbor overall an equal or even higher risk for acute or subacute closure than balloon angioplasty with an acceptably looking result. My tab at the end of the day would have shown one coronary guidewire, one 3.0 mm balloon, and 1–3 DES, rather than 2 coronary guidewires, two balloons, and about 7 or 8 (I lost count) DES. The thriftiness would have carried a risk of about 30% to wind up with one or two take-off restenoses of diagonal branches, unlikely to be clinically apparent. The other risks would have been the same as with the authors approach. The 5 Fr catheter without an introducer could have been pulled at the end, requiring only a 10–15 minute manual compression, rather than a closure device or a mechanical compression device, usually used in conjunction with a 7 Fr guiding catheter inserted through an introducer (about 8.5 Fr outer diameter). I failed to mention the reduced amount of x-ray and contrast medium used, and the shorter occupation of the catheterization laboratory with the simple approach (perhaps even if you include the remote possibility of a redilatation later). Howard Cohen, MD Cardiovascular Institute - UPMCHS Presbyterian University Hospital Pittsburgh, Pennsylvania Bifurcation lesions continue to represent a difficult technical challenge for the interventional cardiologist because of side or main branch compromise due to the problem of plaque shift and the “snowplow effect.” Prior to the advent of stenting, the “kissing balloon” technique or debulking with directional atherectomy was advocated to avoid these problems. With the advent of stenting, attempts to solve this problem with “bifurcation stenting” with the culotte, Y, T or kissing stent techniques were attempted with excellent acute angiographic results and procedural outcomes, but were ultimately abandoned as a primary strategy because of the high rates of restenosis that were encountered. Now, with DES, there is renewed hope that the routine use of these stents in bifurcation lesions will result in superior long-term outcomes. In this issue of the Journal, Dr. Daemon et al. present a dramatic case of the use of DES to treat 3 bifurcations in a patient in a single setting. Eight stents were employed to achieve excellent angiographic results acutely as well as at 6-month follow-up. At a cost of approximately $3,000/stent, this amounts to $24,000 for the stents alone. This is clearly a costly strategy, although it conceivably might be cost effective in the long term if it avoids repeated PCI for restenosis. I suspect that the stents used in this reported case were available on a “research” basis and that cost was not an issue. In our center, as in most cardiac catheterization laboratories worldwide, a “cost”-conscious approach is mandatory, although the needs and safety of the patient are always the foremost consideration. For the patient in this case, I would have tried to avoid the DES in the main circumflex crossing the large marginal branch, particularly as the attempt to recanalize the total occlusion was unsuccessful. Thereafter, a single long stent to cover all the disease in the obtuse marginal and the proximal circumflex could have been employed, thereby avoiding the 2 additional DES that were used. Once the first stent had been placed across the large marginal, the operators in the case report were then committed to a culotte technique to treat the ostium of the large marginal. The LAD and the two diagonals could be treated with a single long stent and kissing balloons in the diagonals as a primary strategy. The angiographic results, however, would undoubtedly be less than ideal. The crush technique, therefore, is potentially an excellent strategy (in regard to short- and long-term outcome) for the bifurcation lesions in the LAD. The two diagonals and the LAD disease are treatable with the crush technique with only three DES. First, two guidewires are inserted into the first and second diagonals, and a third guidewire is inserted into the distal LAD. The first stent is inserted undeployed in the second diagonal. A long, DES is placed undeployed in the LAD covering the first and second diagonals and the diseased segment in the LAD in its entirety. The second diagonal stent is then deployed partially in the LAD and the balloon and wire removed. A DES is then advanced into the first diagonal and deployed partially in the LAD and the balloon and wire removed. Finally, the LAD stent is deployed crushing the diagonal stents that protrude into the LAD and the balloon removed. Both bifurcations can be then hopefully reaccessed and treated with kissing balloons. This approach would allow treatment of the 3 lesions with 4 instead of 8 DES. Admittedly, this is still a costly approach, but hopefully, more cost effective in the long run. The authors of this case are to be congratulated for the excellent results that they achieved both acutely and in the 6-month follow-up. This represents, however, a single case and a large series where at least 1-year outcomes will be necessary to determine whether or not DES overcome the problem of increased restenosis associated with bifurcation stenting. The issues regarding costs will hopefully diminish with newer DES and competitive pricing. David E. Kandzari, MD Director, Interventional Cardiology Research Duke Clinical Research Institute Durham, North Carolina Even in an era of enthusiasm surrounding improved clinical outcomes with DES, catheter-based treatment of bifurcation coronary disease will remain a challenge for interventionalists. As a reflection of these uncertainties, there remains wide variability in operator technique and procedural strategy for bifurcation disease. Irrespective of stent type and coating, coronary angioplasty and stenting in bifurcations is commonly associated with mechanical plaque transfer and shift, contributing to sidebranch occlusion. Further, if the plaque is not excised, the problem of plaque shift often results in an alternating pattern of “chasing” atheromatous material with angioplasty between the parent vessel and sidebranch. Although atherectomy has not consistently demonstrated improved clinical outcomes compared with stenting alone, bifurcation disease appears to be one particular lesion subset in which treatment with directional atherectomy may result in reduced restenosis and target vessel revascularization. Nevertheless, these observations require confirmation in a larger, ongoing clinical study. In this example, my initial preference would be to review the patient’s eligibility for enrollment in the Complex Bifurcation Atherectomy (COMBAT) trial, evaluating a novel catheter for longitudinal directional atherectomy (Foxhollow, Redwood City, California). The catheter is a 6 Fr, guide-compatible, monorail device that performs plaque excision by advancing the cutting window along the lesion length and avoiding a balloon angioplasty effect. However, if I did not achieve an optimal angiographic and IVUS-guided result with DCA, I would have a low threshold to place a DES in the parent vessel, perhaps followed by kissing balloon dilation. To be sure, atherectomy is not suitable for every bifurcation or interventionalist. Consonant with the wide variability in operator technique, alternative approaches include: 1) stenting both the parent and branch vessels (eg, “crush”, Culotte, or T-stenting techniques); 2) stenting of the parent vessel with balloon dilatation in the branch; 3) cutting balloon pre-dilation followed by stenting; and 4) use of novel bifurcating stents, which presently remain under clinical investigation. Yet despite my shared preference to use sirolimus eluting stents, acceptance of a technique that results in three layers of stent applied (“crushed”) to the vessel wall is challenging, since the technique is based more on popularity rather than efficacy data, and may further contribute to the potential for thrombosis. Also, the potential toxicity of multiple drug layers (particularly with cytotoxic agents) should be considered. Regarding the left circumflex lesion, I am uncertain of the benefit of the added complexity and cost in stenting the mid vessel, only to lead to a total occlusion. To date, even with preliminary registry experience with DES in bifurcation disease, stent implantation in the branch vessel has not reliably demonstrated reductions in restenosis compared with DES in the parent vessel and balloon angioplasty treatment of the sidebranch. These early DES observations also imply higher rates of restenosis in bifurcations compared with less complex lesions. When balloon post-dilation in both branches is required following stenting, angioplasty performed in a kissing fashion is necessary, since single vessel treatment may not only shift plaque, but deform stent apposition in the contralateral vessel. Finally, the recognition of bifurcation disease should call attention to the high procedural and clinical risk associated with these lesions, perhaps a reflection of more extensive vascular burden or cardiovascular comorbidity. Previous reports of 30-day MACE and peri-procedural infarction that approximate 30–40% mandate the need for effective antithrombotic therapy and attention to post-procedural risk modification.