Author + information
- Received June 7, 2005
- Revision received August 24, 2005
- Accepted September 8, 2005
- Published online February 7, 2006.
- Marco Valgimigli, MD,
- Patrizia Malagutti, MD,
- Jiro Aoki, MD,
- Héctor M. Garcia-Garcia, MD,
- Gaston A. Rodriguez Granillo, MD,
- Carlos A.G. van Mieghem, MD,
- Jurgen M. Ligthart, BSc,
- Andrew T.L. Ong, MBBS, FRACP,
- George Sianos, MD, PhD,
- Evelyn Regar, MD, PhD,
- Ron T. Van Domburg, PhD,
- Pim De Feyter, MD, PhD,
- Peter de Jaegere, MD, PhD and
- Patrick W. Serruys, MD, PhD⁎ ()
- ↵⁎Reprint requests and correspondence:
Prof. Patrick W. Serruys, Thoraxcenter, Bd-406, Dr Molewaterplein 40, 3015-GD Rotterdam, the Netherlands.
Objectives The purpose of this study was to investigate the long-term clinical and angiographic profile of sirolimus-eluting stent (SES) versus paclitaxel-eluting stent (PES) in patients undergoing percutaneous intervention for left main (LM) coronary disease.
Background The long-term clinical and angiographic impact of SES as opposed to PES implantation in this subset of patients is unknown.
Methods From April 2002 to March 2004, 110 patients underwent percutaneous intervention for LM stenosis at our institution; 55 patients were treated with SES and 55 with PES. The two groups were well balanced for all baseline characteristics.
Results At a median follow-up of 660 days (range 428 to 885), the cumulative incidence of major adverse cardiovascular events was similar (25% in the SES group vs. 29%, in the PES group; hazard ratio 0.88 [95% confidence interval 0.43 to 1.82]; p = 0.74), reflecting similarities in both the composite death/myocardial infarction (16% in the SES group and 18% in the PES group) and target vessel revascularization (9% in the SES group and 11% in the PES group). Angiographic in-stent late loss (mm), evaluated in 73% of the SES group and in 77% of the PES group, was 0.32 ± 74 in the main and 0.36 ± 0.59 in the side branch in the SES group vs. 0.46 ± 0.57 (p = 0.36) and 0.52 ± 0.42 (p = 0.41) in the PES group, respectively.
Conclusions In consecutive patients undergoing percutaneous LM intervention, PES may perform closely to SES both in terms of angiographic and long-term clinical outcome.
Routine drug-eluting stent (DES) implantation, by reducing the need for target vessel revascularization (TVR) and angiographic restenosis, has been recently proposed as the preferred strategy in poor surgical candidates undergoing percutaneous left main coronary artery (LM) intervention (1–3).
The longest average follow-up available for this treatment is currently one year, and whether sirolimus-eluting stent (SES) or paclitaxel-eluting stent (PES) is performing better in these patients is unknown (1–3).
Similarly, even for more conventional lesions, the differential safety and efficacy profile of these two DES options is largely debated (4–8). When taken together, current evidence possibly suggests that in more complex lesion/patient subsets, SES performs better and more safely than PES (4–7,9).
The percutaneous management of LM lesions is a challenging intervention, where bifurcated vessels, extensive wall calcification, and poor hemodynamic tolerance often coexist during treatment.
The purpose of the present study was to investigate, in a high-risk subset of patients undergoing revascularization in a tertiary referral center, the differential long-term impact of SES compared with PES in terms of clinical and angiographic outcome. Intravascular ultrasound analysis has also been carried out at follow-up to quantity neointimal hyperplasia volume.
Study design and patient population
Since April 16, 2002, SES (Cypher; Johnson & Johnson-Cordis, Warren, New Jersey) has been used as a default strategy for every PCI at our institution as part of the Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital (RESEARCH) registry. From the first quarter of 2003, PES (Taxus; Boston Scientific Corp., Natick, Massachusetts) became commercially available, replacing SES as the strategy of choice in every PCI, as part of the Taxus-Stent Evaluated at Rotterdam Cardiology Hospital (T-SEARCH) registry. As a policy, all elective patients presenting with significant (>50% by visual estimation) LM disease, referred to our institution for coronary revascularization, are evaluated both by interventional cardiologists and by cardiac surgeons, and the decision to opt for PCI or surgery is reached by consensus as previously described (1).
From April 16, 2002, to March 6, 2004, a total of 110 consecutive patients were treated exclusively with one or more DES in the LM as part of an elective or nonelective revascularization procedure and constitute the patient population of the present report. Fifty-five patients first received exclusively SES, available at that time in diameters from 2.25 to 3.00 mm, and then 55 patients received PES, available in diameters from 2.25 to 3.5 mm. To ensure comparability between the two study groups, the Parsonnet surgical risk score, based on both clinical presentation profile and comorbidities, and the William Beaumont Hospital simplified scoring system were calculated for each patient (10,11). Nonelective treatment was defined as a procedure carried out on referral before the beginning of the next working day (12).
The protocol was approved by the hospital ethics committee and is in accordance with the Declaration of Helsinki. Written informed consent was obtained from every patient.
Procedures and postintervention medications
All interventions were performed according to current standard guidelines, and, except for the stent utilization, the final interventional strategy was left entirely to the discretion of the operator. Angiographic success was defined as residual stenosis <30% by visual analysis in the presence of Thrombolysis In Myocardial Infarction (TIMI) flow grade 3. All patients were advised to maintain aspirin lifelong, and clopidogrel was prescribed for 6 months in both groups.
End point definitions and clinical follow-up
The occurrence of major adverse cardiac events, defined as death, nonfatal myocardial infarction, or target vessel revascularization, was recorded. Patients with more than one event were assigned the highest ranked event, according to the previous list. End point definitions were previously reported (1). In order to make the clinical follow-up of the two sequential cohorts of patients comparable, clinical outcome of the SES cohort was censored at two years.
Quantitative angiographic and intravenous ultrasound analysis
Quantitative analyses of all angiographic data were performed with the use of edge-detection techniques (CAAS II; Pie Medical, Maastricht, the Netherlands). A value of 0 mm was assigned for the minimal luminal diameter (MLD) in cases of total occlusion at baseline or follow-up. Binary restenosis (BR) was defined as stenosis of >50% of the luminal diameter in the target lesion. Acute gain was defined as the MLD after the index procedure minus the MLD at baseline angiography. Late loss (LL) was defined as the MLD immediately after the index procedure minus the MLD at angiographic follow-up. Quantitative angiographic measurements of the target lesion were obtained in the in-lesion zone (including the stented segment as well as the margins 5 mm proximal and distal to the stent). Intravascular ultrasound (IVUS) analysis was performed after administration of 200 μg of intracoronary nitroglycerin, with an automated pullback at 0.5 mm/s. All IVUS procedures were recorded on VHS videotape, and images were digitized for analysis. A computer-based contour detection was performed with Qurad QCU analysis software (Curad, Wijk Bij Duurstede, the Netherlands) as previously described (13). Intimal hyperplasia volume was calculated as stent volume minus luminal volume. Percentage intimal hyperplasia was defined as intimal hyperplasia volume divided by stent volume.
The sample size was calculated on the assumption that average late loss in the SES and PES group would be around 0.15 mm and 0.35 mm, respectively, based on previous findings. To detect this effect size with a sigma value of 0.3, 85% power, and a type I error (alpha) of 0.05, 32 patients per group were required. Continuous variables are shown as mean ± standard deviation (SD) and were compared using Student unpaired ttest. Categorical variables are presented as counts and percentages and compared with the Fisher exact test. Survival curves were generated by the Kaplan-Meier method, and survival among groups was compared using the log rank test. Cox proportional hazards models were used to assess risk reduction of adverse events. Multivariable analysis, considering all variables reported in Tables 1 and 2⇓with a p value of less than 0.10, was performed to adjust for possible confounders and identify whether the stent received was an independent predictor of adverse events. Probability was significant at a level of <0.05. All statistical tests were two-tailed. Statistical analysis was performed with Statistica 6.1 software (Statsoft, Tulsa, Oklahoma).
Baseline and procedural characteristics
Baseline and procedural characteristics are shown in Tables 1 and 2. The two groups were well matched for all baseline characteristics. In both cohorts, half of the patients presented with a left ventricular ejection fraction of 45% or less or with acute coronary syndrome as indication to treatment, and 9% of the patients presented with severe hemodynamic compromise at entry. The distal LM was overall involved in two-thirds of cases. In the SES group, nominal stent diameter was smaller—reflecting the unavailability of stent bigger than 3.0 mm during the study period—and cumulative stent length tended to be shorter than in the PES group. Bifurcation stenting was equally employed in both groups with a clear preference for T-stenting and for culotte technique in the SES and PES groups, respectively.
Thirty-one patients (56%) in the SES and 25 (45%) in the PES group received intervention in one or more non-LM lesion(s) during index procedure (p = 0.18). Complete revascularization was achieved in 27 (49%) patients in the SES and 36 (65%) in the PES group (p = 0.17).
Overall procedural success was 98%. In one patient receiving SES, presenting with acute MI and shock, a final TIMI flow grade 1 was obtained. An abrupt irreversible occlusion of the circumflex occurred in one PES patient after deployment of a stent in the left main and proximal left anterior descending coronary artery (LAD).
There were no significant differences between the SES and PES groups in the incidence of major adverse cardiac events (MACE) (death, target vessel revascularization, or myocardial infarction [MI]) during the first 30 days (Table 3).Eight deaths occurred in 10 patients presenting with ST-segment elevation acute myocardial infarction and cardiogenic shock at entry. One elective patient, undergoing LM treatment under left ventricular assist device due to end-stage heart disease, died 2 days after for cardiogenic shock, while the second elective patients died for non-cardiovascular reasons after 19 days. No documented thrombotic stent occlusion occurred in the first 30 days or thereafter.
Long-term clinical outcome
After a median follow-up of 660 days (range 428 to 885 days), the cumulative incidence of MACE did not significantly differ in the SES compared to the PES patients (25% vs. 29%, respectively; hazard ratio [HR] 0.88 [95% confidence interval (CI): 0.43 to 1.82]; p = 0.74) (Fig. 1A).The composite death/MI was 16% in the SES and 18% in the PES group (HR 0.95 [95% CI: 0.38 to 2.33]; p = 0.90) (Fig. 1B), and cumulative incidence of TVR was 9% in the SES and 11% in the PES group (HR 0.77 [95% CI: 0.23 to 2.56]; p = 0.66) (Fig. 1C).
In the elective patient population (43 patients in each group), the cumulative incidence of MACE was similar in the SES (21%) and PES groups (25%; HR 0.77 [95% CI 0.32 to 1.8]; p = 0.55). The composite of death/MI was 9% in the SES and 14% in the PES group (HR 0.66 [95% CI 0.19 to 2.3]; p = 0.52), and the need for TVR was 12% in both groups (HR 0.87 [95% CI 0.24 to 3]; p = 0.8).
The cumulative incidence of MACE was similar in patients receiving single-vessel stenting (24% in the SES and 27% in the PES group) and those treated with bifurcation stenting (31%, HR 1.38 [95% CI 0.47 to 4]; p = 0.55; and 33%, HR 1.22 [95% CI 0.45 to 3.3]; p = 0.69; respectively.
After adjustment for nominal stent diameter and total stent length at multivariable Cox regression analysis, SES implantation as opposed to PES failed to emerge as an independent predictor of MACE (HR 0.79 [95% CI 0.5 to 1.5]; p = 0.66). The same remained true after forcing the Parsonnet score into the model.
Thirty-five patients in the SES (73% of eligible patients) and 38 patients in the PES group (77% of those eligible) underwent angiographic follow-up. Data regarding the quantitative coronary angiography for main and side branches are presented in Tables 4 and 5,⇓respectively.
In the SES and PES groups, the main treated branch circumflex was LM-LAD in 22 (63%) and 21 (55%) patients, respectively followed by LM-circumflex in 6 (17%) and 8 (21%), LM alone in 5 (14%) and 6 (16%), and LM-intermediate branch in 2 (6%) and 3 (8%), respectively. In unprotected patients in the SES and PES groups, LM-LAD was the main treated branch in 77% and in 63%, respectively, followed by LM alone in 15% and 17%, LM-circumflex in 8% and 14%, and LM-intermediate branch in 0% and 6%, respectively.
Baseline and follow-up angiographic variables did not differ in the two study groups. No difference was noted in terms of BR in the two groups as the result of a similar in-stent and in-lesion LL (Table 4).
All baseline angiographic variables were well matched between SES and PES groups in the side branches receiving stent. At follow-up, both LL and BR were similar in the two study groups. For those side branches that did not undergo stenting as part of LM treatment, despite a bigger reference vessel diameter in the SES than in the PES group, the pattern of LL was around zero in both groups (Table 5).
When both main and side branches were evaluated on a patient basis, in-lesion BR occurred in 7 (20%) and 12 (32%) patients in the SES and PES groups, respectively (p = 0.44). All cases of nonocclusive restenosis were focal (length <10 mm).
Overall, 46 patients (18 in the SES and 28 in the PES group) underwent IVUS investigation at follow-up. Their baseline and procedural characteristics did not differ from those receiving angiographic examination without IVUS (data not shown). In 11 patients in the SES and 16 in the PES group, LM-LAD was evaluated; the study vessel was LM-CX in 4 SES and 7 PES and LM alone in 3 SES and 5 PES patients. In two patients per group, stent malapposition was noted. As shown in Table 6the degree of neointimal hyperplasia did not differ between the two study groups.
Since their introduction to the market, DES use has steadily increased. Depending on the health care system, they are now partially or almost completely replacing bare-metal stents (BMS) during coronary intervention.
Recently, some concerns for the consequences of using these devices liberally have been raised, which emphasizes the need to scrutinize those patient/lesion subsets that were excluded from landmark randomized trials, particularly beyond conventional eight to nine months’ follow-up (14–17). Yet in both controlled and observational studies, a potential differential efficacy and safety profile between SES and PES has been observed, especially in patient populations considered to be at higher risk for adverse events. Thus, current available evidence reinforces the idea that it would be improper to attribute a class effect to DES and that a high-risk patient population should be better evaluated to further compare the safety/efficacy profile of these two stents.
The percutaneous management of LM lesions is a challenging intervention, where bifurcated vessels, extensive wall calcification, and poor hemodynamic tolerance often coexist during treatment as natural extensions of the anatomic characteristics of the lesions (18,19). Moreover, percutaneous LM intervention, being usually reserved to poor surgical candidates, is often undertaken in patients with low ejection fraction or renal dysfunction, which are known predictors of adverse events even in patients receiving DES (16).
The main results of our analysis show that, as predicted by the risk status of the patients, the overall event rate was higher than that previously reported for non-LM lesions. However, the safety/efficacy profile of the two DES evaluated was apparently maintained at long-term follow-up, with PES performing closely to SES in terms of both clinical and angiographic outcome.
This statement is based on the following findings:
1. The great majority of events occurred in both groups within one year, considering either the whole (86% in the SES and 81% in the PES) or the elective population (78% in the SES and 73% in the PES group). No early or late angiographically confirmed stent thrombosis has been observed, with only one sudden death occurring in an 86-year-old woman affected by a hematological malignancy seven months after the index procedure.
2. The short- and long-term clinical event rate in the two study groups was not different, and at multivariable analysis the stent implanted failed to emerge as an independent predictor of adverse events.
3. Angiographic and IVUS investigation demonstrated that late loss and neointimal hyperplasia volume were similar in the two groups of patients. The angiographic outcome of the nonstented side branches was also remarkably similar between the two groups.
Among these observations, the last one was unexpected and deserves special attention.
In all major randomized controlled trials evaluating the benefit of SES versus BMS, the average in-stent LL for SES was reported to be constantly equal to or below 0.20, whereas the same figure for PES, based on PES versus BMS studies, was around two times higher. The Prospective, Randomized, Multi-Center, Comparison of the Cypher Sirolimus-Eluting and the Taxus Paclitaxel-Eluting Stent Systems (REALITY) study showed a difference between these two DES in terms of angiographic end points that was even bigger, with a mean in-stent LL of 0.9 in SES and 0.31 in the PES group (5). Conversely, in the current investigation we failed to show any difference between the two stents in terms of in-stent or in-segment LL or BR, either in the main branch or in the stented side branches. Interestingly, this was confirmed by the IVUS analysis. Although the difference was more striking for SES, the LL in both groups were actually much higher than the figures expected. The explanation for this discrepancy can only be speculative for the moment and it may suggest that angiographic response to DES is lesion/patient-specific. Data so far available in the literature are inconclusive in this regard. Park et al. (2) recently reported an average late loss of 0.05 mm in 102 patients receiving SES for the treatment of unprotected left main disease, whereas the same figure in 85 LM patients at higher risk status, treated with either PES or SES, was cumulatively reported to be 0.58 mm (3). Unfortunately, in that study no distinction between SES and PES was made (3).
The lack of availability of SES sizes bigger than 3.0 mm during the study period, which imposed an aggressive overdilation strategy to match LM reference diameter, might have theoretically played a role. However, this technical issue was at least partially encountered in the series reported by Park et al. (2) as well.
Alternatively, this difference between studies could possibly reflect a selection bias, with patients at higher clinical risk based on previous cardiovascular history and comorbidities being more prone to develop a more aggressive intimal proliferation after DES. To investigate this possibility in an exploratory fashion (this analysis was not prespecified for the current study), we pooled the LL for both main branch and all stented side branches as the outcome variable in a linear regression model. At univariate analysis including all variables reported in Tables 1 and 2we found protected status to be the strongest predictor for high LL (β = 0.47 [95% CI 0.24 to 0.7]; p = 0.001; adjusted R2= 0.39). Accordingly, we observed that in patients receiving unprotected LM intervention (n = 59) LL was cumulatively lower (0.31 ± 0.41 vs. 0.63 ± 72; p = 0.037) than in patients receiving protected treatment (n = 14), and, of note, all occlusive binary restenosis occurred in the protected group of patients.
Interestingly, none of the variables reported in Tables 1 and 2differed significantly between patients receiving protected versus unprotected LM intervention, with the overall Parsonnet score being 16 ± 7.5 in protected versus 18 ± 10 in unprotected patients (p = 0.4).
The observed difference in LL between protected versus unprotected LM intervention might possibly outline a role for shear stress as potential modulator of vessel response to DES, as recently suggested by our group (20). This analysis was exploratory in nature and clearly beyond the scope of the current investigation. However, it underscores the need to consider DES performance in the context of the patient population in which the device was actually tested.
Taken together, our observations suggest that DES may perform more effectively in good than in poor surgical candidates, which reinforces the interest in assessing the value of this treatment as compared with conventional surgical approach in a properly designed randomized trial.
The present study is a single-center experience from a tertiary referral center and lacks the clear advantages of a multicenter randomized study. In particular, despite the fact that the study was conducted over a relatively short period, we cannot exclude the possibility that improvements in technique or differences in drug prescription could have partially confounded our main results.
Accordingly, the results of our study are encouraging, but they cannot be conclusive. Studies with bigger sample sizes and more prolonged clinical follow-ups are clearly required to rule out the occurrence of less common device-related side effects.
Clinical implications of the combined RESEARCH and T-SEARCH analysis
In the overall results of the unselected RESEARCH versus T-SEARCH comparison, a shift toward more complex lesions has been noted from SES to PES cohort (8). This difference was not confirmed in our current analysis, which focused on LM lesions cumulatively treated over a longer period of time: more patients in the SES group received concomitant intervention in non-LM lesions and fewer reached complete revascularization than in the PES cohort. However, the stent length was greater in the PES group. When this finding is combined with the observed shift from T-stenting in the SES period to culotte in the PES period as bifurcation technique, it may suggest that operators have become progressively more familiar with DES over time and that full lesion coverage with DES had been more frequently performed in the PES than in the SES group. The impact of these confounders on our final result remains incompletely understood.
When taken together, the combined RESEARCH and T-SEARCH analysis may reinforce the concept that in an uncontrolled setting such as our clinical practice, the coronary device in itself should probably be regarded among the principal but clearly not as the only component of the long-term procedural success in the DES era. Rather, the two drug-eluting stents available on the market should always be put in the context of the characteristics of the treated patients and the operator’s experience to better forecast their effect on short- and long-term outcome.
After a median follow-up of two years, no late serious adverse events, possibly suggesting a time-dependent change in the therapeutic profile of the investigated devices, were observed.
In a consecutive group of patients undergoing percutaneous LM intervention, PES may perform closely to SES in terms of both clinical and angiographic outcome.
A multinational multicenter randomized study is currently ongoing to estimate the clinical value of PES-supported LM intervention with respect to conventional surgical treatment.
This study was supported by the Erasmus Medical Center, Rotterdam, and by unrestricted institutional grants from Boston Scientific Corporation (Natick, Massachusetts) and Cordis, a Johnson & Johnson company (Warren, New Jersey).
- Abbreviations and Acronyms
- bare-metal stent
- binary restenosis
- left circumflex coronary artery
- drug-eluting stent
- intravascular ultrasound
- left anterior descending coronary artery
- late loss
- left main coronary artery
- major adverse cardiac events
- myocardial infarction
- minimal luminal diameter
- paclitaxel-eluting stent
- right coronary artery
- sirolimus-eluting stent
- target vessel revascularization
- Received June 7, 2005.
- Revision received August 24, 2005.
- Accepted September 8, 2005.
- American College of Cardiology Foundation
- Valgimigli M.,
- van Mieghem C.A.,
- Ong A.T.,
- et al.
- Park S.J.,
- Kim Y.H.,
- Lee B.K.,
- et al.
- Chieffo A.,
- Stankovic G.,
- Bonizzoni E.,
- et al.
- De Lezo J.,
- Medina A.,
- Pan M.
- ↵Morice M. Eight-month outcome of the REALITY study: a prospective, randomized, multicenter head-to-head comparison of the sirolimus eluting stent (Cypher) and the paclitaxel eluting stent (Taxus). Paper presented at: 2005 Scientific Session of the American College of Cardiology; March 6, 2005; Orlando, FL.
- Ong A.T.,
- Serruys P.W.,
- Aoki J.,
- et al.
- Qureshi M.A.,
- Safian R.D.,
- Grines C.L.,
- et al.
- Roques F.,
- Nashef S.A.,
- Michel P.,
- et al.
- Hamers R.,
- Knook M.,
- Roelandt J.
- Waksman R.,
- Kuchulakanti P.,
- Kinnaird T.,
- et al.
- Alfonso F.,
- Moreno R.,
- Vergas J.
- Tan W.A.,
- Tamai H.,
- Park S.J.,
- et al.
- Takagi T.,
- Stankovic G.,
- Finci L.,
- et al.