Incomplete Stent Apposition in a Left Main Bifurcated Lesion after Kissing Stent Implantation

The introduction of drug-eluting stents in clinical practice has markedly lowered the incidence of coronary restenosis,^1,2 although the management of bifurcation lesions remains challenging.³ The “kissing” stent technique, whereby stents placed in the main and side branch from the proximal segment of the main branch are inflated simultaneously, was originally proposed to facilitate and expedite the treatment of left main coronary artery (LMCA) bifurcation lesions with bare-metal stents.⁴ Since it allows complete coverage of the lesion and optimal stent expansion, this technique, which is applicable to bifurcations with narrow angles and to lesions of both ostia without involvement proximal to the bifurcation, is now being reevaluated for use with drug-eluting stents.^5–11 However, it has several limitations, including the need for a wide guiding catheter, the placement of a metallic neocarina in the main branch that might cause subacute or late thrombosis and difficult reaccess to each branch.^5–9 We encountered a case of restenosis of the ostium of the left circumflex (LCx) artery after kissing drug-eluting stent implantation was performed at the LMCA bifurcation. The relationship between stent characteristics (expansion and deformation) and clinical restenosis after kissing stent implantation has not been clearly defined. We therefore investigated the stent overlap, distortion and apposition to the vessel using a phantom three-dimensional LMCA bifurcation model, and also reexamined the stent apposition in the ostium of the LCx by fluoroscopic and intravascular ultrasound (IVUS) imaging. Case Report. A 75-year-old female was admitted to our hospital for evaluation of recurrent exercise-induced angina and a positive treadmill test. Her coronary risk factors included hypertension, diabetes mellitus, hyperlipidemia, obesity and status post-menopause. She had sustained an acute myocardial infarction 4 months earlier, and underwent a percutaneous revascularization procedure with implantations of Driver™ coronary stents (Medtronic, Inc., Minneapolis, Minnesota) in the proximal (3.5 x 15 mm) and mid (3.0 x 24 mm) left anterior descending (LAD) artery. She underwent further myocardial revascularization 3 weeks later for > 90% stenoses of the proximal and mid-LCx arteries with 2.5 x 13 mm MultiLink Penta™ (Guidant Corp., Indianapolis, Indiana) and 2.5 x 15 mm Tsunami™ (Terumo, Tokyo, Japan) stents, respectively. A 4-month follow-up coronary angiogram revealed the presence of: (a) a 75% stenosis of the distal LMCA; (b) 90% proximal and 75% mid-LAD in-stent restenoses; and (c) 75% proximal and 90% mid-LCx in-stent restenoses (Figures 1 A–C). After explaining risks and benefits for both treatment options, the patient decided against surgery and opted for interventional treatment. Informed consent was obtained from the patient. An 8 Fr, JL 4.0 guiding catheter was inserted into the left coronary artery and floppy wires were advanced in the LAD and the LCx arteries. A 2.9 Fr Atlantis Pro™ IVUS catheter (Boston Scientific Corp., Natick, Massachusetts), advanced into each artery, showed intimal proliferation at each stent implantation site and demonstrated that the LMCA, proximal LAD and proximal LCx arteries were 5.0 mm, 3.8 mm, and 3.7 mm in vessel diameter (distance from media to media), respectively. After predilatation of each lesion, a 2.5 x 23 mm Cypher™ sirolimus-eluting stent (SES) (Cordis Corp., Miami, Florida) was implanted in the mid-LCx before the insertion from the distal LMCA of 3.5 x 23 mm and 3.0 x 28 mm SES in the LAD and LCx arteries, respectively, with overlapping of the two LCx stents. The two SES at the LMCA bifurcation were then simultaneously inflated at 12 atm (Figure 1D). While maintaining the stent balloons in the same positions, the LCx, LAD, and both balloons simultaneously, were sequentially inflated at 18, 18 and 12 atm, respectively. Post-stenting angiography confirmed the satisfactory dilatation of each stenotic lesion, and showed the presence of an intimal flap in the LMCA, proximal to the site of stent implantation (Figures 1E and F). IVUS imaging showed a 6.6 mm2 minimal area of the SES in the proximal LAD (Figure 2A), and a double-barrel image in the distal LMCA (Figure 2B). Within the double barrel, the LAD stent was dilated to 3.6 mm in its long axis, 2.7 mm in its short axis and 9.1 mm2 in area, while the LCx stent was compressed to 1.6 mm in its short axis, with the area preserved at 4.9 mm2. The overall barrel was oval in shape, 4.3 mm in its long axis and 3.6 mm in its short axis. A 0.9-mm hematoma was observed just proximal to the LAD stent and was associated with a small intimal tear and a luminal area preserved at 10.9 mm2 (Figure 2C). No further intervention was performed. Although the patient’s CPK enzyme peaked to 412 IU/ml on the next day, she did not develop any chest pain or any prominent electrocardiographic changes. Follow-up coronary angiograms performed 6 months later showed a 25% stenosis of the LMCA at the site of the hematoma, and an isolated 75% stenosis of the LCx ostium at the site of the LCx and LAD SES overlap (Figure 3). Since the patient had suffered no angina, medical treatment was continued. She has remained free of adverse clinical events over a 12-month follow-up period. In Vitro Testing with a Three-Dimensional Model. In vitro testing of kissing stent implantation using a three-dimensional model was performed to investigate: (a) the effect of the difference in stent sizes on stent expansion at the double-barrel portion; and (b) the impact of stent overlap at the distal LMCA on stent apposition to the vessel wall. Methods and Materials. A bifurcated vinyl tubing was attached along the LMCA, LAD and LCx on an acrylic three-dimensional human heart model. The following kissing stent implantations using Multilink Penta™ stents were performed, varying the method of stent implantation or the stent size: (a) deployment of 3.5 x 23 mm and 3.0 x 28 mm stents in the LAD and LCx arteries from the distal LMCA, respectively, positioning the LAD stent over the LCx stent (Figure 4A); (b) deployment of 3.5 x 23 mm and 3.0 x 28 mm stents, respectively, positioning the LCx stent over the LAD stent (Figure 4B); (c) deployment of 3.0 x 28 mm and 3.0 x 23 mm stents, respectively, positioning the LAD stent over the LCx stent (Figure 4C); and (d) deployment of 3.0 x 28 mm and 3.0 x 28 mm stents, respectively, positioning the LCx stent over the LAD stent (Figure 4D). The kissing stent technique was the same as that used during the clinical case, i.e., the 2 stents at the LMCA bifurcation were simultaneously deployed at 12 atm. While maintaining the stent delivery balloons in the same position in the LCx and LAD, both balloons were simultaneously inflated at 18, 18 and 12 atm, sequentially. Results. Effect of stent size on stent expansion. In the in vitro testing using different size stents, the small-size stents (3.0 mm) were compressed by the large-size stents (3.5 mm) at the double-barrel portion (Figures 4 A and B). In the experiments using same-size stents, both stents were expanded to the round shape, and obvious stent distortion was not observed (Figures 4 C and D). The double-barrel portion was extremely expanded when the 2 stents were deployed simultaneously in all variations. Impact of stent overlapping on the apposition to the vessel. Incomplete stent apposition in the vessel tubing was most prevalent beneath the overlapping stent (Figure 5). However, there was no significant difference in the degree of stent distortion or in the area of incomplete stent apposition created between the two types of stent overlap (Figure 4). Discussion. Kissing stent implantation is a quick and feasible technique that obtains simultaneous expansion of both vessels of the bifurcation. This technique was already studied during the bare-metal stent era, with reports of acceptable long-term results.⁸ There was a decrease in in-hospital and 30-day major adverse cardiac events (MACE) rates in the kissing stent group when compared to the provisional T-stenting group. The 9-month target lesion revascularization rate was also decreased in the kissing stent group (5% versus 18% in the provisional T-stenting group).⁸ Sharma et al also reported that the target lesion revascularization rate was reduced to 4% in 200 patients who underwent kissing stent implantation using SES, with no evidence of late thrombosis.⁹ Park et al reported 17 cases with kissing stent implantations using SES at the LMCA bifurcation.¹¹ He demonstrated 3 cases (18%) of angiographic restenosis and 2 cases (12%) of target lesion revascularization at 12 months.¹¹ Kissing stent implantation was selected to treat our patient for the following reasons: (a) the patient had a large and patent proximal LMCA, with the lesion from the distal LMCA to both the LAD and LCx ostia; and (b) since a diffuse, severe restenotic lesion found in the bare-metal stents involved both proximal sites of the LAD and LCx, this area needed to be completely covered by SES. We concluded that T-stenting was not suitable for this patient due to the gap that would be created at the ostium of the side branch³ and an increased target lumen revascularization rate compared with crush stenting.¹² We also concluded that crush stenting was not suitable because her LMCA was too large to be covered by a maximally expanded SES. It has been reported that incomplete stent apposition can occur at the orifice of the side branch after crush stenting in right-angled bifurcation cases.¹³ In this case, the LMCA was injured by deploying 2 parallel stents at the proximal site of the bifurcation. Stents 3.5 mm and 3.0 mm in diameter are both predicted to expand to 4.6 mm in diameter and 16.7 mm2 in area by the kissing balloon inflation technique according to the formula proposed by Mitsudo: R2 = r12 + r22, where R is the proximal main branch diameter, and r1 and r2 are the diameters of the bifurcated branches.¹⁴ The actual stent area, however, was smaller than predicted and not round, but oval. The presence of a small lesion in the LMCA, and the stiffness of the SES limited the stent expansion. The LCx stent was compressed by the LAD stent despite simultaneous inflation to 12 atm. The LCx balloon risked being overdilated to form a “dog-bone” shape, causing a proximal dissection. Similar stent distortion was observed in the in vitro models using 3.5 mm and 3.0 mm stents (Figures 4 A and B), while it was not observed in the models using two 3.0 mm stents (Figures 4 C and D). This result suggests that the selection of the same size stent in this technique would prevent proximal injury and compression of the LCx stent, although IVUS imaging shows that the LAD has a larger vessel diameter. When an injury or restenosis occurs to the proximal site with this technique, reintervention is a serious issue.^5,6 Two methods of intervention can be considered. The first method involves adding another drug-eluting stent, while the other method involves detaining 2 drug-eluting stents in the form for addition to the double barrel.^8,9 The first method presents some issues such as the difficulty in not making the gap between the stents, and in possibility transforming the metallic carina. The issue with the second method is an increased risk of thrombosis due to an extended metallic carina. After crushing the side branch stent, deployment of another drug-eluting stent at the more proximal site of the main branch could also be a feasible solution. There has been little reported about restenosis caused by stent overlapping and distortion at the bifurcated area after kissing stent procedures, while incomplete stent apposition to the bifurcated vessel is thought to be a main mechanism of restenosis after crush stenting.^12,13,15,16 Although it has been demonstrated that incomplete stent apposition at the ostium of the side branch occurs after crush stenting,^13,15 additional kissing balloon inflation has been reported to improve stent apposition and reduce restenosis at the side branch by 60–70%.^12–16 In the in vitro tests, one stent was always positioned over the other stent in the distal LMCA, with either the LAD stent positioned over the LCx stent, or vice versa. One cause of stent overlap may be the different curvatures in each vessel. The overlap of the stents produced an incomplete stent apposition beneath the overlap at the orifice of the branch (Figure 5), which is suggested to be one of the mechanisms of restenosis after kissing stent implantation. Although there were no significant differences in the area of incomplete stent apposition, stent expansion and distortion of the overlapping stents, the plaque burden at the bifurcation should be evaluated by IVUS prior to deployment. Careful consideration of positioning and stent overlap should be made to avoid creating a gap at the site of significant plaque. In our patient, at the 6-month follow-up angiography, a restenotic lesion was observed at the LCx ostium. The restenotic lesion had developed in the area where the LCx and LAD stents overlapped. The overlapping stents were confirmed by fluoroscopy (Figures 3 D–I) and also by IVUS. Intimal injury was observed on the LCx side of the LMCA in the angiographic images (Figure 1 F) and at the 12 o’clock position of the IVUS image (Figure 2 C). In a more distal IVUS image (Figure 2B), the LAD stent was visualized at the 12 o’clock position where the LCx artery originated, demonstrating stent crossover at the bifurcation. The overlap of the stents may have produced a gap beneath the LCx stent at the orifice of the LCx (Figure 6). The incomplete apposition of the LCx stent due to its deformation at the bifurcation is another considerable mechanism of restenosis. The in vitro experiment suggests that the difference in stent size is attributed to the deformity of the side branch stent and deviation of the metallic carina (Figures 4 A and B). It has been reported that the BX Velocity stent, the stent platform for the SES, is less conformable in right-angled vessel curvature,¹⁷ and the stent bending may be occurring at the wide-angled bifurcation. Study limitations. Since observations made in the disease-free vinyl tubing may not necessarily apply to actual clinical cases with fibrotic and calcified vessels, the stent distortion or gap formation may be emphasized in the model. We used the Multi-Link Penta™ stent in this in vitro model; however, there may be some differences in flexibility, conformability or stiffness of the stent compared with other stent platforms for drug-eluting stents. Conclusion. In summary, we encountered a restenosis case after kissing SES stent deployment at the LMCA bifurcation. We studied the characteristics of kissing stent implantation in a three-dimensional LMCA bifurcation model to clarify the mechanism of restenosis. It was demonstrated that the stent overlap at the distal LMCA created an incomplete stent apposition beneath the overlapping stent, and that using different-sized stents produced compression of the LCx stent at the distal LMCA. These phenomena were confirmed with fluoroscopy and IVUS in our clinical case. The incomplete stent apposition induced by these phenomena is thought to be the primary mechanism of restenosis after kissing stent implantation.

References 1. Morice MC, Serruys PW, Sousa JE, et al for the RAVEL Study Group. Randomized study with the sirolimus-coated Bx Velocity balloon-expandable stent in the treatment of patients with de novo native coronary artery lesions. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med 2002;346:1773–1780. 2. Moses JW, Leon MB, Popma JJ, et al. for the SIRIUS Investigators. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315–1323. 3. Colombo A, Moses JW, Morice MC, et al. Randomized study to evaluate sirolimus-eluting stents implanted at coronary bifurcation lesions. Circulation 2004;109:1244–1249. 4. Colombo A, Gaglione A, Nakamura S, Finci L. "Kissing" stents for bifurcational coronary lesion. Cathet Cardiovasc Diagn 1993;30:327–330. 5. Melikian N, Airoldi F, Di Mario C. Coronary bifurcation stenting. Current techniques, outcome and possible future developments. Minerva Cardioangiol 2004;52:365–378. 6. Lefevre T, Louvard Y, Morice MC, et al. Stenting of bifurcation lesions: A rational approach. J Interv Cardiol 2001;14:573–585. 7. Iakovou I, Ge L, Colombo A. Contemporary stent treatment of coronary bifurcations. J Am Coll Cardiol 2005;46:1446–1455. 8. Sharma SK, Choudhury A, Lee J, et al. Simultaneous kissing stents (SKS) technique for treating bifurcation lesions in medium-to-large size coronary arteries. Am J Cardiol 2004;94:913–917. 9. Sharma SK. Simultaneous kissing drug-eluting stent technique for percutaneous treatment of bifurcation lesions in large-size vessels. Catheter Cardiovasc Interv 2005;65:10–16. 10. Cox N, Rogers C. Kissing sirolimus-eluting stents for the treatment of left main coronary artery stenosis. Catheter Cardiovasc Interv 2004;61:206–210. 11. Park SJ, Kim YH, Lee BK, et al. Sirolimus-eluting stent implantation for unprotected left main coronary artery stenosis: comparison with bare metal stent implantation. J Am Coll Cardiol 2005;45:351–356. 12. Ge L, Iakovou I, Cosgrave J, et al. Treatment bifurcation lesions with two stents: Crush versus T Stenting — One year angiographic and clinical follow-up. Heart 2005, June 17 Epub. 13. Ormiston JA, Currie E, Webster MW, et al. Drug-eluting stents for coronary bifurcations: Insights into the crush technique. Catheter Cardiovasc Interv 2004;63:332–336. 14. Mitsudo K. PTCA Technique (2nd Edition). Igakusyoin:Tokyo, Japan. 2001, pp. 133–140 (in Japanese). 15. Murasato Y, Suzuka H, Kamezaki F. Vascular endoscopic and macroscopic observations after crush stenting of coronary artery bifurcations in pigs. Catheter Cardiovasc Interv 2005;66:237–243. 16. Ge L, Airoldi F, Iakovou I, et al. Clinical and angiographic outcome after implantation of drug-eluting stents in bifurcation lesions with the crush stent technique: Importance of final kissing balloon post-dilation. J Am Coll Cardiol 2005;46: 613–620. 17. Rieu R, Barragan P, Garitey V, et al. Assessment of the trackability, flexibility, and conformability of coronary stents: A comparative analysis. Catheter Cardiovasc Interv 2003;59:496–503.