Stent Thrombosis With Second- Versus First-Generation Drug-Eluting Stents in Real-World Percutaneous Coronary Intervention: Analysis of 3806 Consecutive Procedures From a Large-Volume Single-Center Prospective Registry

Abstract: Background and Aims. When compared to their first-generation (1^stGEN) counterparts, second-generation (2^ndGEN) drug-eluting stents (DESs) have been associated with better clinical outcomes in randomized clinical trials, namely by reducing the rates of stent thrombosis (ST). Our goal was to investigate whether or not the broad use of newer devices would translate into higher safety in a real-world population. For that purpose, we compared the occurrence of definite ST at 12 months between two patient subsets from a large-volume single-center registry, according to the type of DES used. Total mortality was a secondary endpoint. Methods and Results. Between January 2003 and December 2010, a total of 3806 patients were submitted to percutaneous coronary intervention (PCI) with only 1^stGEN or 2^ndGEN DES: 2388 patients (62.7%) were treated with 1^stGEN DES only (sirolimus-eluting stent [SES] = 1295 [34.0%]; paclitaxel-eluting stent [PES] = 943 [24.8%]; both stent types were used in 150 patients) and 1418 patients (37.3%) were treated with 2^ndGEN DESs only. The total incidence of definite ST (as defined by the Academic Research Consortium) at 12 months was 1.2% (n = 46). After correction for baseline differences between study groups and other variables deemed to influence the occurrence of ST, the use of 1^stGEN DES was associated with a significant 2.4-fold increase in the risk of definite ST (95% confidence interval [CI], 1.05-5.42; P=.039) at 12 months; adjusted risk was higher with PES (hazard ratio [HR], 3.6; 95% CI, 1.48-8.70; P=.005) than with SES (HR, 2.3; 95% CI, 0.92-5.65; P=.074). Total mortality (3.7% vs 3.5%) did not differ significantly between groups (adjusted HR, 1.2; 95% CI, 0.81-1.84, P=.348). Conclusions. Our data suggest that in the real-world setting of contemporary PCI, the unrestricted use of newer 2^ndGEN DESs translates into an improvement in PCI safety (relative to 1^stGEN DESs), with a significantly lower risk of definite ST at 12 months.

J INVASIVE CARDIOL 2013;25(7):330-336

Key words: stent thrombosis, drug-eluting stent

____________________________________________

Stent thrombosis (ST) is a serious and often fatal event limiting the efficacy of percutaneous coronary intervention (PCI). The pathophysiology of ST is multifactorial, and underlying causes including stent-, procedure-, lesion-, and patient-related factors seem to play different roles at different time points after the index procedure.^1,2 When compared to first-generation (1^stGEN) drug-eluting stents (DESs), newer DESs have been associated with a lower rate of ST in several randomized clinical trials, subsequent meta-analyses, and also in some registries, such as the recently published Swedish Coronary Angiography and Angioplasty Registry (SCAAR).^3-7 New, second-generation (2^ndGEN) DESs have been developed with improved design and materials, both of which may contribute to overcome some of the limitations of the older DESs. Decreased strut thickness — resulting in higher flexibility, conformability, and deliverability — and optimized polymer biocompatibility and drug delivery kinetics have been shown to contribute to a low late-loss rate and to a lower thrombotic risk.¹ Despite the evidence pointing in this direction, most of the data comes from post hoc analysis and meta-analysis, mainly because studies defining ST as a primary endpoint are scarce. 

We aimed to assess whether or not the systematic use of a 2^ndGEN DES, relative to the 1^stGEN DES, translates into a higher safety rate in a real-world population where DES implantation was indicated. For that purpose, we conducted an analysis of a single-center prospective registry, evaluating the incidence of definite ST, as defined by the Academic Research Consortium (ARC), at 12 months of follow-up as the primary outcome measure.

Methods

Study population. Between January 2003 and December 2010, all consecutive PCI procedures performed at our institution with implantation of at least 1 DES (n = 8172) were initially screened. Cases in which a bare-metal stent, any combination of 1^stGEN and 2^ndGEN DESs (according to study criteria), or with isolated balloon angioplasty (including treatment of in-stent restenosis) were excluded. Subsequently, study groups were defined according to whether only 1^stGEN or only 2^ndGEN DESs were implanted (Figure 1). Sirolimus-eluting  stents (SESs) and paclitaxel-eluting stents (PESs) were classified as 1^stGEN; this group included Cypher, Cypher Select, and Cypher Select Plus stents (Cordis Corporation) as well as Taxus Express and Taxus Liberté stents (Boston Scientific Corporation). Everolimus-eluting stents (EESs), ie, XienceV and Xience Prime stents (Abbott Laboratories) and Promus and Promus Element stents (Boston Scientific Corp.), and zotarolimus-eluting stents (ZESs), ie, Endeavor, Endeavor Sprint, and Endeavor Resolute stents (Medtronic, Inc), were classified as 2^ndGEN DESs. In the case of staged procedures or non-target lesion revascularization (TLR), subsequent interventions in the same patient were included as an “index procedure” if and only if the initial inclusion criteria would still hold true. The final population consisted of 3806 procedures performed purely with 1^stGEN (n = 2388) or 2^ndGEN (n = 1418) DESs, in which a complete 12-month follow-up was available. The study flow chart is depicted on Figure 1. Patients were included in the single-center Angiography and Coronary Revascularization Registry of Santa Cruz Hospital (ACROSS) registry, in which demographic, clinical, angiographic, and procedure-related variables are prospectively collected using a dedicated lab-based computer database (Cardiobase, Infortucano). 

Revascularization procedures. Choice of treatment strategy was at the physician’s discretion for each individual case. All patients underwent PCI using standard interventional techniques, in accordance with our department’s protocol at each time point, and the type of stent used and the steps performed for angiographic optimization were according to the physician’s preference. Antiplatelet and anticoagulant therapy followed international standard recommended regimens. Long-term information on dual-antiplatelet therapy compliance and duration was not available. The SYNTAX score was calculated using a computed algorithm included in our database (retrospectively for the procedures performed before 2009). Procedure success was defined as final Thrombolysis in Myocardial Infarction (TIMI) flow III and residual stenosis <30% by visual estimation after stent implantation.

Endpoints and follow-up. The primary endpoint of the study was the occurrence of ST at 12 months, according to the ARC definition.⁸ All-cause mortality was analyzed as an isolated secondary measure during the same period. For the events that occurred before 2007, the ARC criteria were retrospectively adjudicated. Primary endpoint adjudication was performed by angiographic analysis and confirmed for every case recorded in the database by reviewing the qualifying angiograms. Angiographic follow-up was not mandatory and was decided individually according to physician preference and patient clinical status. Thus, cases of ST were diagnosed following a clinically indicated coronary angiography, usually resulting from an acute coronary syndrome (ACS); vital status was accessed by clinical file review, telephone interview, or (when no other information was available) using the national Social Security database. Study time frames were chosen to allow a complete 12 months of follow-up for all patients treated with 2^ndGEN stents by the time the analysis was performed.

Statistical analysis. Continuous variables with normal distribution were expressed as means and standard deviations. Normality was tested with the Kolmogorov-Smirnov test and/or by visual assessment of Q-Q plots. Discrete variables were expressed as frequencies and percentages. Statistical comparison of baseline characteristics and outcomes was performed using the Chi-square test or Fisher’s exact test, when appropriate, for categorical variables and the Student’s t-test or the Mann-Whitney test for continuous variables. All performed analyses were retrospective. As stated before, inclusion was per procedure, meaning that the same patient may have contributed more than once to the total risk estimates; however, only the first event was adjudicated in the ST analysis. Corrected risk estimates were performed using a Cox proportional hazard regression model. The following variables were included as covariates: age, gender, previous myocardial infarction, hypertension, smoking, diabetes, acute coronary syndrome in the index procedure, PCI of a bypass graft or treatment of stent restenosis, number of treated vessels per procedure, maximal stent inflation pressure, minimal stent diameter, and SYNTAX score. Hazard ratio (HR) for all-cause mortality according to stent type was adjusted for age, gender, diabetes, coronary syndrome in the index procedure, and SYNTAX score. Two-tailed tests of significance are reported. For all comparisons, a P-value of <.05 was considered statistically significant. When appropriate, 95% confidence interval (CI) was calculated. Statistical analysis was performed with SPSS version 19.0 (SPSS, Inc).

Results

Population characteristics and clinical setting. During the study period, a total of 3844 PCI procedures using exclusively first- or second-generation DESs were performed. Thirty-eight patients (0.99%) were lost to follow-up; 3806 procedures were finally eligible for primary endpoint and survival analysis. Two hundred forty-seven patients (6.5%) had a repeat PCI as part of a staged procedure or as a non-TLR new coronary revascularization. In the majority of cases (62.7%), a 1^stGEN SES was implanted; ZESs were the most frequently used 2^ndGEN DES (Figure 1). As expected, during the first 3 to 4 years of the registry, there was a clear predominance of 1^stGEN DESs, mainly because newer devices were not commercially available at that time and their use was infrequent. Only after 2007 did the use of both 1^stGEN and 2^ndGEN DESs become roughly similar and truly contemporary.

Baseline patient characteristics are depicted in Table 1. Briefly, index procedures with 2^ndGEN DESs were performed in older patients (65.4 ± 11.1 years vs 63.2 ± 10.6 years; P<.001), with significantly higher prevalence of hypertension (77.8% vs 72.2%; P<.001), lower prevalence of current/former smoking (37.4% vs 48.6%; P<.001), and more often in the setting of an ACS (48.7% vs 41.4%; P<.001). Patients treated with 1^stGEN DESs were more frequently male (76.0% vs 73.0%; P=.041), and had a higher prevalence of previous myocardial infarction (30.2% vs 27.1%; P=.044) and previous myocardial revascularization, both percutaneous (43.3% vs 30.6%; P<.001) and surgical (16.8% vs 11.9%; P<.001).

Angiographic and procedural characteristics. Angiographic and procedure-related characteristics are shown in Table 2. Overall, patients treated with 2^ndGEN DESs had a lower burden and complexity of coronary artery disease, as reflected by the SYNTAX score, which was slightly but significantly lower (13.9 ± 10.3 vs 13.1 ± 9.7; P=.010). Also, the number of treated vessels per procedure and maximal inflation pressure were lower, and PCI of a bypass graft or stent restenosis as an indication for the intervention in the index procedure occurred less often in the 2^ndGEN group.

Primary endpoint analysis. During the study-defined 12-month follow-up period, forty-six cases of definite ST were recorded in the entire cohort (total incidence, 1.2%). Overall, 1^stGEN DES patients had a significantly higher absolute incidence of 1.6%, compared to 0.6% in 2^ndGEN DES patients (P=.005). The median time between the index PCI and ST was 43 days (range, 4-167 days), and was not significantly different between the two study groups (P=.118). Figure 2 shows the total incidence of ST according to treatment year and DES type. In 2003, 2004, and 2005, almost all procedures were performed using 1^stGEN DESs (Figure 3). During this period, the rates of definite ST were remarkably high, especially in 2003. When considering only the time frame during which the use of both stent generations was contemporary (starting in 2006), the difference in the absolute incidence of the primary endpoint, although still numerically higher in 1^stGEN DES patients (1.2% vs 0.6%), was no longer statistically significant (P=.083). The exclusion of patients with repeat procedures yielded results that were consistent with the baseline model. 

Importantly, from the total of ST cases, 69.6% (n = 32) occurred in patients treated in the setting of an ACS (P<.001) and the majority (n = 24; 52.2%) were late ST (occurring between 30 days and 12 months). For all time periods (Figure 4), according to the ARC consensus, the incidence of the primary endpoint was always numerically superior in the 1^stGEN group, but a statistically significant difference was detected only in the case of late ST: acute ST was 0.3% vs 0.1% (P=.104), subacute ST was 0.4% vs 0.2% (P=.289), and late ST rate was 0.8% vs 0.3% (P=.036). 

In order to correct for the potential effect of differences in baseline characteristics and of other variables known to influence the likelihood of stent thrombosis, a Cox proportional hazard logistic regression model was used; variables included as covariates were listed above. In the resulting prediction model, implantation of a 1^stGEN DES was independently associated with a significant 2.38-fold higher risk of definite ST at 12 months (95% CI, 1.05-5.42; P=.039) (Figure 5A). When 1^stGEN DESs were analyzed separately and compared to any 2^ndGEN DES, only PESs were significantly associated with the occurrence of definite ST at 12 months, with a corrected HR of 3.59 (95% CI, 1.4-8.7; P=.005). The point estimate for the HR for 1^stGEN SES still favored 2^ndGEN devices, but was not statistically significant (HR, 2.28; 95% CI, 0.92-5.64; P=.074) (Figure 5B).

Other predictors of ST at 12 months were PCI in the setting of an ACS in the index procedure (HR, 3.27; 95% CI, 1.05-5.42; P=.001) and the SYNTAX score (HR, 1.03 per unit increase; 95% CI, 1.0-1.05; P=.039); PCI of a bypass graft or treatment of stent restenosis was associated with a non-significant trend toward an higher risk of the primary endpoint (HR, 1.9; 95% CI, 0.93-4.03; P=.08). 

Secondary endpoint analysis. Concerning the 12-month all-cause mortality, a total of 138 events were registered, with a similar incidence between the two study groups (3.7% 1^stGEN vs 3.5% 2^ndGEN; P=.8). After correction for age, gender, and ACS at presentation, implantation of 1^stGEN DES was not significantly associated with all-cause mortality at 12 months (HR,1.04; 95% CI, 0.74-1.48; P=.8) (Figure 6).

Discussion

In this large, single-center registry of a real-world all-comers population, we showed that relative to first 1^stGEN DESs, the unrestricted use of 2^ndGEN DESs is associated with a significantly lower incidence of definite ST at 12 months after implantation, especially in the case of late ST. Despite being a potentially catastrophic complication of coronary intervention that can result in acute myocardial infarction and death, we could not document a significant difference in the all-cause mortality rate at 12 months between study groups. 

Concerning the cumbersome issue of ST, randomized trials upon which PCI recommendations are based have been underpowered and were not specifically designed to detect differences in this rare endpoint, except for the noteworthy exception of the recently published PROTECT trial.⁹ In addition, the inconsistency of the definitions used in the many trials make direct comparisons difficult and ultimately led the ARC to orchestrate a set of consensus definitions for DES study endpoints.⁸ 

Several randomized clinical trials and registries have compared different generations of DESs and have been published in the last few years. Overall, their results are consistent with our findings. In the recently published SCAAR Registry,² a significant reduction in the rate of definite ST was associated with the implantation of newer DESs, when compared to older devices. Importantly, our cohort had both more complex coronary disease and an overall higher baseline risk profile, and in the SCAAR Registry,² the Endeavor Sprint CrCo ZES was classified as old (1^stGEN DES). In the Western Denmark Heart Registry¹¹ and in the Bern-Rotterdam Registry,¹² first-generation PESs had a higher rate of ST than SESs; again, this is in line with the results from the present study. In the latter, however, there was also a significant difference in the 12-month all-cause mortality. On the contrary, in the ESTROFA Registry,¹³ ST was not statistically different between PES and SES groups at 3 years of follow-up. 

On the other hand, randomized trials have produced mixed results, in that some inconsistencies exist concerning the reported rates of definite ST at 12 months according to the type of stent: higher in PES versus EES (SPIRIT II¹⁴, III¹⁵, and IV¹⁶, COMPARE¹⁷), higher in PES versus ZES (ENDEAVOR IV¹⁸, ZEST¹⁹), higher in SES versus EES (SORT OUT IV²⁰), higher in ZES versus SES (ZEST²¹, SORT OUT III²²), similar in ZES versus SES (PROTECT⁹), similar in ZES versus EES (RESOLUTE all-comers²³), and similar in SES versus PES (SIRTAX²⁴, LATE²⁵). In a recently published network meta-analysis of 49 randomized trials including 50844 patients, the second-generation CrCo EES was shown to have lower ST rates as compared to all other commercially available common-use drug-eluting and even bare-metal stents.²⁶ In these trials, as in our registry, PESs have been consistently associated with worse clinical outcomes, including higher rates of ST. These comparatively poor results of PES may have several explanations, including drug release kinetics, the drug itself, the polymer, or other intrinsic characteristics of the stent, which are likely to predispose to ST.^27-29 In our cohort, when considered separately, both of the studied 1^stGEN DES types, PES and SES, were associated with a higher rate of definite ST when compared to 2^ndGEN DESs, but the corrected HR was significantly higher for PESs only.

Study limitations. Despite being in consonance with previously published literature, our results need to be interpreted with caution, because the study has some important limitations. Some of the observed differences in baseline characteristics between study groups could have influenced the results. The angiographic burden of coronary artery disease differed significantly between groups, favoring the 2^ndGEN DES patients; for this reason, the SYNTAX score was included in the regression model and was an independent predictor of ST. Importantly, patients presenting with an ACS at the index procedure were more frequent in the 2^ndGEN DES subgroup. This was most likely related to the increase in the proportion of acute unstable events as the indication for PCI that has been consistently observed over time. However, since ACS is a well-known risk factor for ST^27,28 (and a predictor of the occurrence of the primary endpoint in our analysis, independent of the type of stent used), if any, this unbalanced distribution would have had a negative effect on the 2^ndGEN group. Correction for other clinical and procedure-related characteristics that have previously been associated with ST, such as diabetes, treatment of in-stent restenosis, PCI of a graft, minimal stent diameter ,and maximal inflation pressure have also been accounted for by including these variables in the regression model. As previously noted, not surprisingly, during the first triennium of the enrollment period, more than 90% of the procedures were performed using 1^stGEN DESs, simply because newer devices were not available for widespread use. The Endeavor CrCo ZES became legally available in Europe in September 2005 and others soon followed. The incidence of total ST was not uniform during the study period and a trend toward a reduction in its incidence could be detected. During the first few years of DES use (after their debut in 2002), the enthusiastic results regarding strikingly low restenosis rates relative to bare-metal stents^30,31 may have led to excessive “off-label” use of DESs, which in turn could have contributed to the exceedingly high rates of ST that were observed in the early period (mainly, 2003 and 2004). Additionally, improvements in the technique and increased awareness of the need for PCI optimization when using 1^stGEN DESs may also have influenced these results. In unpublished data, we were able to show in our population that a significant increase in the peak stent inflation pressure and postdilatation use occurred over time during the DES era. Additionally, new DES devices are deployed through more compliant balloon systems, rendering stent undersizing and malapposition less frequent. Overcoming these limitations, despite adequate statistical corrections, is an inherent limitation of retrospective designs. Changes in PCI technique over time most likely could not have been optimally captured by the variables included in the prediction model and thus the actual weight of their contribution to the results could not be firmly assessed. In keeping with these observations, when only the period during which the use of both stent types was truly contemporary, the absolute difference in 12-month ST rates — although still numerically higher in 1^stGEN DES-treated patients — was smaller and no longer statistically significant.

According to the ARC definition of time of occurrence, the rate of definite ST was numerically higher for 1^stGEN DESs in all studied periods; although a statistically significant difference could not be found, a trend was evident regarding the rate of late ST, also favoring 2^ndGEN DESs.

Given the relatively infrequent occurrence of ST, large populations are required to detect small differences in event rates. Under-reporting of ST cases cannot be overcome with retrospective designs. Although the event rate in our dataset is in accordance with other reports, the follow-up was limited to 12 months to allow an even comparison between groups, which makes estimation of very-late ST risk impossible. The absence of data related to antiplatelet therapy adherence and compliance during the follow-up period and, importantly, the absence of information from intravascular ultrasound and other PCI adjunctive therapies/interventions, such as thrombectomy, glycoprotein IIb/IIIa inhibitors, and stent postdilatation, make it impossible to assess the impact of these important variables on the reported outcomes. Despite the use of appropriate statistical adjustments, some residual confounders could have remained.

Conclusion

Our registry suggests that in a real-world setting, patients treated with 1^stGEN DESs have a higher risk of definite ST at 12 months after implantation. Despite the fact that the absolute difference in definite ST rates appeared to attenuate over time, the authors feel that the systematic use of the newer second-generation devices, whenever a DES may be indicated, is associated with a significant improvement in PCI safety by reducing the rates of definite ST.

References

1. Holmes DR Jr, Kereiakes DJ, Garg S, et al. Stent thrombosis. J Am Coll Cardiol. 2010;56(17):1357-1365.
2. Van Werkum JW, Heestermans AA, Zomer AC, et al. Predictors of coronary stent thrombosis. The Dutch Stent Thrombosis Registry. J Am Coll Cardiol. 2009;53(16):1399-1409.
3. Sarn G, Lagerqvist B, Frobert O, et al. Lower risk of stent thrombosis and restenosis with unrestricted use of ‘new-generation’ drug-eluting stents: a report from the nationwide Swedish Coronary Angiography and Angioplasty Registry (SCAAR). Eur Heart J. 2012;33(5):606-613.
4. Planer D, Smits PC, Kereiakes DJ, et al. Comparison of everolimus and paclitaxel-eluting stents in patients with acute and stable coronary syndromes: pooled results from the SPIRIT (A Clinical Evaluation of the XIENCE V Everolimus Eluting Coronary Stent System) and COMPARE (A Trial of Everolimus-Eluting Stents and Paclitaxel-Eluting Stents for Coronary Revascularization in Daily Practice) trials. JACC Cardiovasc Interv. 2011;4(10):1104-1115.
5. Cassese S, Piccolo R, Galasso G, et al. Twelve-month clinical outcomes of everolimus-eluting stent as compared to paclitaxel- and sirolimus-eluting stent in patients undergoing percutaneous coronary interventions. A meta-analysis of randomized clinical trials. Int J Cardiol. 2011;150(1):84-89.
6. de Waha A, Cassese S, Park DW, et al. Everolimus-eluting versus sirolimus-eluting stents: an updated meta-analysis of randomized trials. Clin Res Cardiol. 2012;101(6):461-467.
7. Stone GW, Newman W, Mastali K, et al. Everolimus-eluting versus paclitaxel-eluting stents in coronary artery disease. N Engl J Med. 2010;362(18):1663-1674.
8. Cutlip D, Windecker S, Mehran R, et al. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation. 2007;115(17):2344-2351.
9. Kamenzind E, Wijns W, Mauri L, et al. Stent thrombosis and major clinical events at 3 years after zotarolimus-eluting or sirolimus-coronary stent implantation: randomized, multicenter, open-label, controlled trial. Lancet. 2012;380(9851):1396-1405.
10. Kolandaivelu K, Swaminathan R, Gibson W, et al. Stent thrombogenicity early in high-risk interventional settings is driven by stent design and deployment and protected by polymer-drug coatings. Circulation. 2011;123(13):1400-1409.
11. Kaltoft A, Jensen LO, Maeng M, et al. 2-year clinical outcomes after implantation of sirolimus-eluting, paclitaxel-eluting, and bare-metal coronary stents: results from the WDHR (Western Denmark Heart Registry). J Am Coll Cardiol. 2009;53(8):658-664.
12. Daemen J, Wenaweser P, Tsuchida K, et al. Early and late coronary stent thrombosis of sirolimus-eluting and paclitaxel-eluting stents in routine clinical practice: data from a large two-institutional cohort study. Lancet. 2007;369(9562):667-678.
13. Torre-Hernandez J, Alfonso F, Hernandez F, et al. Drug-eluting stent thrombosis results from the multicenter Spanish registry ESTROFA (Estudio ESpañol sobre TROmbosis de stents FArmacoactivos). J Am Coll Cardiol. 2008;51(10):986-990.
14. Serruys PW, Ruygrok P, Neuzner J, et al. A randomised comparison of an everolimus-eluting coronary stent with a paclitaxel-eluting coronary stent: The SPIRIT II trial. EuroIntervention. 2006;2(3):286-294.
15. Stone GW, Midei M, Newman W, et al. Randomized comparison of everolimus-eluting and paclitaxel-eluting stents: two-year clinical follow-up from the Clinical Evaluation of the Xience V Everolimus Eluting Coronary Stent System in the Treatment of Patients with de novo Native Coronary Artery Lesions (SPIRIT) III trial. Circulation. 2009;119(5):680-686.
16. Stone GW. SPIRIT IV. Late breaking clinical trial. Presented at Transcatheter Cardiovascular Therapeutics, San Francisco, California, September 23, 2009.
17. Smits P. COMPARE trial. Presented at Transcatheter Cardiovascular Therapeutics, San Francisco, California, September 23, 2009.
18. Leon M. The ENDEAVOR and ENDEAVOR Resolute zotarolimus-eluting stent: comprehensive update of the clinical trial program (featuring the first presentation of the ENDEAVOR IV 3-year results). Paper presented at Transcatheter Cardiovascular Therapeutics, San Francisco, California, September 21, 2009.
19. Park DW, Kim YH, Yun SC, et al. Comparison of zotarolimus-eluting stents with sirolimus and paclitaxel-eluting stents for coronary revascularization. The ZEST (Comparison of the Efficacy and Safety of Zotarolimus-Eluting Stent with Sirolimus-Eluting and PacliTaxel-Eluting Stent for Coronary Lesions) randomized trial. Am Coll Cardiol. 2010;56(15):1187-1195.
20. Jensen LO, Thayssen P, Hansen HS, et al. Randomized comparison of everolimus-eluting and sirolimus-eluting stents in patients treated with percutaneous coronary intervention: the Scandinavian Organization for Randomized Trials with Clinical Outcome IV (SORT OUT IV). Circulation. 2012;125(10):1246-1255.
21. Park DW, Kim YH, Yun SC, et al. Comparison of zotarolimus-eluting stents with sirolimus and paclitaxel-eluting stents for coronary revascularization. The ZEST (comparison of the efficacy and safety of zotarolimus-eluting stent with sirolimus-eluting and paclitaxel-eluting stent for coronary lesions) randomized trial. J Am Coll Cardiol. 2010;56(15):1187-1195.
22. Rasmussen K, Maeng M, Kaltoft A, et al. Efficacy and safety of zotarolimus-eluting and sirolimus-eluting coronary stents in routine clinical care (SORT OUT III): a randomized controlled superiority trial. Lancet. 2010;375(9720):1090-1099.
23. Serruys PW, Silber S, Garg S, et al. Comparison of zotarolimus-eluting and everolimus-eluting coronary stents.N Engl J Med.2010;363(2):136-146.
24. Raber L. SIRTAX-LATE: five-year clinical and angiographic  follow-up from a prospective randomized trial of sirolimus-eluting and paclitaxel-eluting stents. Presented at Transcatheter Cardiovascular Therapeutics, San Francisco, California, September 22, 2009.
25. Räber L, Wohlwend L, Wigger M, et al. Five-year clinical and angiographic outcomes of a randomized comparison of sirolimus-eluting and paclitaxel-eluting stents: results of the sirolimus-eluting versus paclitaxel-eluting stents for coronary revascularization LATE trial. Circulation. 2011;123(24):2819-2828.
26. Palmerini T, Biondi-Zoccai G, Della Riva D, et al. Stent thrombosis with drug-eluting and bare-metal stents: evidence from a comprehensive network meta-analysis. Lancet.2012;379(9824):1393-1402. 
27. Joner M, Finn AV, Farb A, et al. Pathology of drug-eluting stents in humans. J Am Coll Cardiol. 2006;48(1):193-202.
28. Cayla G, Hulot JA, O’Connor S, et al. Clinical, angiographic, and genetic factors associated with early coronary stent thrombosis. JAMA. 2011;306(16):1765-1774.
29. Chitkara K, Gershlick A. Second versus first-generation drug-eluting stents. J Interv Cardiol. 2010;5:23-26.
30. Morice MC, Serruys PW, Eduardo SJ, et al; the RAVEL study group. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med. 2002;346(23):1773-1780.
31. Moses JW, Leon MB, Popma JJ, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med. 2003;349(14):1315-1323.

 _______________________________

From the Department of Cardiology, Hospital de Santa Cruz – Centro Hospitalar de Lisboa Ocidental, Lisbon, Portugal.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Teles is a consultant and grantholder for Abbott and a consultant for Medtronic. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript submitted January 11, 2013, provisional acceptance given February 6, 2013, final version accepted March 20, 2013.

Address for correspondence: Dr Helder Dores, Cardiology Department – Santa Cruz Hospital, Centro Hospitalar Lisboa Ocidental, Av. Professor Reinaldo dos Santos, 2799-523 Carnaxide, Lisbon, Portugal. Email: heldores@hotmail.com