Unprotected Left Main Coronary Artery Bifurcation Stenosis

Impact of Plaque Debulking Prior to Single Sirolimus-Eluting Stent Implantation

ABSTRACT: Background. The impact of plaque debulking with directional coronary atherectomy (DCA) prior to single sirolimus-eluting stent (SES) implantation in an unprotected left main coronary artery (LMCA) involving bifurcation stenosis has not been fully evaluated.  Methods. One hundred and one patients with unprotected LMCA bifurcation lesions treated with single SES implantation (from the LMCA to the left descending coronary artery [LAD] across the left circumflex artery [LCx] ostium) were divided into 2 groups: DCA group (n = 41, plaque debulking with DCA prior to SES implantation) and non-DCA group (n = 60, single SES implantation alone). Clinical outcomes as well as angiographic data at baseline, post procedure and follow up were compared between the two groups. Results. At 1-year follow up, freedom from major adverse cardiac events was 97.4 ± 2.6% in the DCA group, and 88.6 ± 4.4% in the non-DCA group (p = 0.129). Baseline quantitative coronary angiographic analyses revealed that the percent diameter stenosis (%DS) of the LCx was higher in the DCA group than in the non-DCA group (36.8 ± 21.5% vs. 26.9 ± 19.2%; p = 0.029). Moreover, the %DS of the LCx after PCI and at 9-month follow up was lower in the DCA group (19.2 ± 13.1% vs. 28.3 ± 22.7%; p = 0.034; 20.8 ± 12.3% vs. 31.9 ± 21.4%; p = 0.007, respectively). Furthermore, restenosis at the LCx ostium was not observed in the DCA group, but was seen in 5 cases in the non-DCA group (0% vs. 10.2%; p = 0.048). Conclusion. Plaque debulking with DCA prior to single SES implantation effectively reduced restenosis of the LCx ostium in this challenging lesion subset.

J INVASIVE CARDIOL 2008;20:505–510

Treatment of unprotected left main coronary artery (LMCA) stenosis by percutaneous coronary intervention (PCI) continues to represent a considerable challenge for contemporary interventionalists. Even in the bare-metal stent (BMS) era, coronary artery bypass grafting (CABG) was often the recommended treatment.1 However, current availability and application of drug-eluting stents (DES) have provided a more favorable clinical outcome by decreasing in-stent restenosis in the treatment of unprotected LMCA disease.2–4

There are several reports about PCI efficacy using sirolimus eluting-stent (SES) implantation for unprotected LMCA stenosis, and the results differ according to the exact lesion location.4–7 Park et al reported that the overall restenosis rate in the LMCA involving bifurcation lesions was higher than in the body or ostium of LMCA lesions (9.8% vs. 0%), with restenosis occurring mainly at the left circumflex artery (LCx) ostium.4 Therefore, one issue in the treatment of LMCA bifurcation stenosis is how to prevent restenosis of the LCx ostium. According to a recent report, single-stent treatment strategy offered a better outcome compared with dual-stent strategies.6

Recently, Tsuchikane et al reported on the efficacy of plaque debulking using directional coronary atherectomy (DCA) for bifurcation lesions prior to SES implantation.8 We hypothesized that pre-stent adjunctive DCA could potentially improve the therapeutic outcome of LMCA bifurcation stenosis intervention. Thus, the aim of this study was to evaluate the impact of plaque debulking with DCA prior to single SES implantation for unprotected LMCA bifurcation lesions.

Methods

Study population. Between June 2004 and September 2006, 151 consecutive patients with de novo unprotected LMCA lesions, excluding ST-elevation myocardial infarction, underwent PCI with SES (Cypher™, Cordis Corp., Miami Lakes, Florida) at Toyohashi Heart Center (Figure 1). An unprotected LMCA was defined as having no patent coronary artery bypass grafts to the left anterior descending artery (LAD) or left circumflex artery (LCx). The lesion was located at the bifurcation in 123 patients of the 151 patients, at the ostium in 16 patients and at the body in 12 patients. Twenty-two patients underwent SES implantation at both the LAD and LCx ostium using the dual-stent strategies and were therefore excluded. The remaining 101 patients, where a single SES was implanted from the LMCA to the LAD across the LCx ostium were eligible for this study. The subjects were divided into two groups: the DCA group (n = 41, plaque debulking with DCA prior to single SES implantation) and the non-DCA group (n = 60, single SES implantation alone). Clinical outcomes during the 1-year follow up as well as angiographic data at baseline, post procedure and at follow up were compared between the two groups.

Procedure and medication. After local anesthesia, an 8 Fr sheath was placed in the femoral artery. Intravenous heparin was given to achieve an activated clotting time > 300 seconds. In the DCA group, plaque debulking was performed using a 7 Fr Flexi-Cut L device (Abbott Vascular, Abbott Park, Illinois). All patients underwent an intravascular ultrasound (IVUS)-guided DCA procedure in the LMCA and LAD, and if there was significant narrowing at the LCx ostium, DCA was performed to the LCx at the operator’s discretion. DCA of the LCx was performed in 14 of 41 patients in the DCA group. Plaque debulking was attempted to achieve a residual percent plaque plus media cross-sectional area < 60%. In the non-DCA group, predilatation was performed in 54 of 60 patients. In both groups, a SES was implanted to cover the lesion fully according to standard clinical practice, and high-pressure inflation was performed to achieve optimal stent expansion and apposition, namely a lumen cross-sectional area (CSA) > 90% of the distal reference lumen CSA.9,10 Final kissing balloon postdilatation after SES implantation was also performed in both groups. All patients received oral aspirin (100 mg/day) indefinitely, and oral ticlopidine (200 mg/day) or clopidogrel (75 mg/day) for at least 3 months. Glycoprotein IIb/IIIa antagonists were not used in either group, as these agents are not licensed in Japan.

Quantitative coronary angiographic (QCA) analysis. All baseline, procedural, and follow-up angiograms were performed immediately after administration of 200 µg of intracoronary nitroglycerin, and the treated lesion was evaluated using two or more angiographic projections. QCA was performed by two experienced angiographers using the Cardiovascular Measurement System (CMS-MEDIS, Medical Imaging Systems, Leiden, Netherlands). Lesion length, reference diameter, minimal lumen diameter (MLD), and % diameter stenosis (%DS) were measured using the view showing the smallest luminal diameter at diastolic frames. The analysis was performed in the main branch (from the LMCA to the LAD) and in the side branch (LCx) at baseline, post-procedure and a 9-month follow up.

Definitions. Acute gain was defined as the difference between baseline and postprocedural MLD, and late loss was defined as the difference between postprocedural and follow-up MLD. Procedural success was defined as achievement of < 30% angiographic residual stenosis and thrombolysis in myocardial infarction (TIMI) flow grade 3 in both the main and side branch,11 without periprocedural or in-hospital complications such as death, Q-wave myocardial infarction (QMI), or emergent bypass surgery. QMI was defined according to biochemical or electrocardiographic criteria: the MB isoform of creatine kinase (CK-MB) at least 3 times the upper limit of the normal range in at least 1 blood sample, or the finding of abnormal Q-waves in ≥ 2 contiguous leads. Angiographic restenosis was defined as a diameter stenosis > 50% within the stented segment plus the 5 mm proximal and distal persistent area at follow-up angiography. Target lesion revascularization (TLR) was defined as any revascularization or bypass surgery performed because of angiographic restenosis in association with angina pectoris and objective evidence of myocardial ischemia. Major adverse cardiac events (MACE) were defined as the occurrence of cardiac death, myocardial infarction (MI), TLR in-hospital and during the follow-up period.

Patient follow up. Clinical patient evaluations were performed at regular intervals. Repeat coronary angiography was scheduled 9 months after the index procedure unless clinical reasons such as symptoms or documentation of myocardial ischemia warranted earlier intervention. MACE, including cardiac death, MI and target vessel revascularization was assessed at 1 year after SES implantation.

Statistical analysis. Statistical analysis was performed using Stat-View, version 5 (SAS Institute, Inc., Cary, North Carolina). Data were expressed as mean ± 1 SD for continuous variables and as frequencies for categorical variables. The chi-square test or Fisher’s exact test was used for comparing frequencies of occurrence, and the unpaired t-test for continuous variables. A p-value < 0.05 was considered statistically significant.

Results

Patient and procedural characteristics. Baseline patient characteristics and clinical data for both groups are shown in Table 1. Baseline patient characteristics were similar, except for more male subjects in the DCA group than in the non-DCA group. No cases of coronary perforation or major coronary dissection during the DCA procedure were observed. Neither stent thrombosis nor QMI occurred during hospitalization in the DCA group. Finally, procedural success was achieved in all patients in the DCA group. A QMI occurred in 1 patient in the non-DCA group due to acute side branch (LCx) occlusion. There were no incidents of death or emergent bypass surgery during hospitalization in either group.

Clinical follow up. All patients were followed regularly for more than 1 year. The mean clinical follow-up period was 15.6 ± 7.1 months in the DCA group, and 17.6 ± 8.2 months in the non-DCA group. In addition, there was no significant difference in dual antiplatelet therapy compliance between the two groups (aspirin and ticlopidine or clopidogrel; 92.7% in the DCA group, 93.3% in the non-DCA group) during follow up. Cumulative clinical events at 1 year after the procedure are shown in Table 2. At 1 year, the event-free MACE rate was 97.4 ± 2.6% in the DCA group and 88.6 ± 4.4% in the non-DCA group (p = 0.129) (Figure 2). During the follow-up period, 1 patient (2.4%) in the DCA group underwent TLR at the proximal stent edge of the LMCA. Five patients (8.3%) in the non-DCA group underwent TLR: 2 patients for both the LMCA and LCx, 1 for the LMCA, and 2 for the LCx. During follow up, death occurred in 4 patients. One patient died of respiratory disease (at 10.9 months) in the DCA group. In the non-DCA group, 1 patient died of cancer (at 4.9 months); 1 died of rupture of an abdominal aortic aneurysm (at 13.1 months); and 1 patient with poor left ventricular ejection fraction died of ventricular fibrillation during hemodialysis (at 4.7 months).

Angiographic characteristics. Medina’s classification of bifurcation lesions12 is summarized in Table 3 and quantitative coronary angiographic data of the LMCA are shown in Table 4. Nine-month follow-up angiography was performed in 36 patients (87.8%) in the DCA group and 49 patients (81.7%) in the non-DCA group. Baseline angiographic variables of LMCA were similar for both groups. Postprocedural MLD of the LMCA was larger and the %DS was lower in the DCA group than in the non-DCA group (3.71 ± 0.66 mm vs. 3.24 ± 0.57 mm; p < 0.001; 9.0 ± 9.1% vs. 15.6 ± 9.3%; p < 0.001, respectively). Furthermore, these differences were maintained at 9-month follow up. Acute gain was larger in the DCA group than in the non-DCA group (2.34 ± 0.61 mm vs. 1.88 ± 0.59 mm; p < 0.001), but late loss in each group was similar (0.38 ± 0.46 mm vs. 0.29 ± 0.47 mm; p = 0.344). The angiographic binary restenosis rate of the stented segment was 2.7% in the DCA group and 6.1% in the non-DCA group (p = 0.472). No aneurysmal formation was observed in either group.

Quantitative coronary angiographic data on the non-stented orifice of the LCx are shown in Table 5. Baseline %DS was higher in the DCA group than in the non-DCA group (36.8 ± 21.5% vs. 26.9 ± 19.2%; p = 0.029), nevertheless postprocedural MLD was larger and the %DS was lower in the DCA group, compared to the non-DCA group (2.42 ± 0.56 mm vs. 2.09 ± 0.72 mm; p = 0.023, 19.2 ± 13.1% vs. 28.3 ± 22.7%; p = 0.034, respectively). These differences were maintained at 9-month follow up. While acute gain was larger in the DCA group than in the non-DCA group (0.65 ± 0.78 mm vs. 0.14 ± 0.75 mm; p = 0.003), late loss was similar (0.13 ± 0.24 mm DCA vs. 0.15 ± 0.67 mm non-DCA; p = 0.826, respectively). No restenosis of the LCx orifice occurred in the DCA group, but did occur in 5 patients (10.2%) in the non-DCA group (p = 0.048). Serial assessment of %DS in both the LMCA and LCx are shown in Figures 3 A and B.

Discussion

With the advent of DES and the resultant dramatic reduction in restenosis rates, there has been overwhelming enthusiasm for its application in more complex lesion subsets including LMCA lesions. There are several reports regarding SES implantation for LMCA lesions.4–7 According to these reports, SES implantation would be safe and effective for lesions located at the body or ostium of the LMCA, but clinical results are less favorable for lesions involving the LMCA bifurcation.

Recently, numerous strategies for bifurcation stenting using 2 stents, including T-stenting, culotte stenting, kissing stenting and crush stenting, have been developed. However, the restenosis rates for the side branch were still higher in comparison to those of the main branch, ranging between 13.5 and 25.3%.6,13–16 Using IVUS data, a preliminary analysis of dual-stent strategies demonstrated that incomplete stent expansion at the side branch ostium17 and incomplete dispersion of the coated drug could be significant causes of restenosis.

Previous data on bifurcation stenting suggest that complex dual-stent strategies were associated with relatively higher restenosis rates of the side branch compared to a single-stent strategy.6,14 Furthermore, repeat intervention might be difficult when in-stent restenosis occurs in the lesion treated with 2 stents. Taken together, the single-stent strategy should be preferred for bifurcation lesions. However, in cases with severe lesions in the ostium of the side branch, especially the LCx ostium in the LMCA involving a bifurcation, single-stent strategy could be ineffective in preventing acute side branch occlusion or restenosis at the side branch ostium. A recent report has shown that plaque debulking with DCA for bifurcation lesions before SES implantation was effective in the prevention of plaque shift to the side branch ostium, and it suggested the possibility of plaque debulking as an alternative to complex dual-stent strategies.8 Our data demonstrate that this strategy was associated with the prevention of procedural side branch occlusion and reduction of residual stenosis at the ostium of the LCx in spite of more severe stenosis of the ostium at baseline compared to the non-DCA group. Adequate plaque debulking would minimize the significant plaque shift, which is the most important cause of acute side branch occlusion or residual severe stenosis post procedure. Furthermore, it is also notable that the difference in residual stenosis post procedure between both groups was maintained at follow up. Possible mechanisms to account for these favorable effects of this technique include not only prevention of plaque shift, but also mechanical distention of the vessel wall.

These favorable effects would result from appropriate plaque debulking. For this purpose, IVUS should play an important role in the procedure to confirm plaque distribution and in decision-making of endpoints.18 With IVUS-guided DCA procedures, most series reported a high procedural success rate.19 In fact, there were no complications seen in the periprocedural period of the DCA group in this study. The MACE-free survival rate at 1-year follow up also showed a higher trend in the DCA group than in the non-DCA group.

It is still controversial whether PCI or CABG offers better outcomes in the treatment of the LMCA involving a bifurcation. Compared to recent reports of LMCA stenosis treated with CABG,20,21 our results suggest the possibility that DCA prior to single SES implantation in unprotected LMCA bifurcation lesions could be as feasible as CABG. On the other hand, PCI using IVUS-guided DCA makes the procedure more complicated and time-consuming and requires an 8 Fr delivery system. Therefore, development of new convenient debulking devices with a smaller profile is recommended.

Study limitations. First, this study was a retrospective, observational study. There was no randomization and the DCA procedure was performed at the operator’s discretion. Second, the sample size was relatively small. These limitations prevent reaching a definitive conclusion as to the efficacy of plaque debulking. A large, randomized, multicenter clinical study is warranted to more accurately evaluate this interventional approach.

Conclusion

This study showed that IVUS-guided plaque debulking with DCA prior to single SES implantation for an unprotected LMCA involving a bifurcation stenosis was safe and effective. Although the restenosis rate of the LMCA was not affected, the restenosis rate of the LCx ostium was significantly reduced.

Acknowledgments. The authors wish to thank Heidi N. Bonneau, RN, MS, CCA, and Hideaki Kaneda, MD, for their expert review of the manuscript.