We compared the incidence of late clinical events after withdrawal of clopidogrel between subjects treated with drug-eluting stents (DES) and those treated with bare-metal stents. Death and myocardial infarction occurred more frequently among DES-treated subjects during the follow-up period. The results of this study indicate that there may be a penalty for the lower rate of restenosis and reinterventions after DES implantation, in particular, an increased rate of late stent thromboses.
The use of drug-eluting stents (DES) can significantly reduce the rate of restenoses and target vessel revascularizations (TVR) compared with bare-metal stents (BMS).1,2 Stent thrombosis is a complication that occurs less frequently than restenosis, but it is coupled with greater morbidity and mortality.3,4 For this reason, there is growing concern that the occurrence of late stent thromboses caused by delayed endothelialization or hypersensitivity reactions may be more likely with the use of DES.5
Randomized studies initially revealed a similar incidence of stent thrombosis for both DES and BMS implantation (0.5%-0.6%); however, the observation periods were a maximum of 6 to 12 months after stent implantation.6 It has been shown in case studies that stent thromboses may occur more than 1 year after DES implantation.7 Withdrawal of dual-antiplatelet therapy especially appears to increase the risk of stent thromboses.7,8 We performed the Basel Stent Kosten-Effektivitats-Late Thrombotic Events (BASKET-LATE) study to compare the incidence of late clinical events after withdrawal of clopidogrel (Plavix) between DES-treated and BMS-treated subjects.9
Subjects and methods
The 826 consecutive subjects who had participated in the BASKET study10 were included in the BASKET-LATE trial. In BASKET, subjects were randomly assigned to treatment with DES (Cypher Cordis, Johnson & Johnson Corporation, Miami, Fla; or Taxus, Boston Scientific Corporation, Natick, Mass) or BMS (Vision, Guidant Corporation, Indianapolis, Ind) in a 2:1 manner. The BASKET-LATE study included all 746 subjects who survived the first 6 months without myocardial infarction (MI) or TVR, with a total of 1133 stented lesions. Subjects were followed for an additional 12 months to determine "late" clinical events.
Clinical events were defined as cardiac death or nonfatal MI. Late was defined as onset between 7 and 18 months after stent implantation. Sudden cardiac death and nonfatal MI that could be ascribed to the target vessel were classified as "thrombus related." Target vessel revascularizations not associated with thrombosis-related events were considered "restenosis related." The combined end point "major cardiac events" consisted of cardiac death, nonfatal MI, and TVR.
All subjects were treated for 6 months with 100 mg aspirin and 75 mg clopidogrel daily. Clopidogrel therapy was withdrawn after 6 months, but aspirin was continued long term. The drug therapy, therefore, was the same for both treatment groups.
We observed 6 cardiac deaths and 23 nonfatal MIs following withdrawal of clopidogrel therapy after 6 months. All deaths and 20 of the 23 nonfatal MIs occurred in subjects with DES implantation. As shown in the Table, the combined end point of cardiac death and nonfatal MI occurred more frequently in the period 7 to 18 months after stent implantation in subjects with DES compared with BMS implantation (4.9% vs 1.3%, respectively; .01). On the other hand, there was a tendency toward fewer restenosis-related TVR in DES-treated subjects (4.5% vs 6.7%, respectively; .21).
Table. Late major cardiac events (months 7 to 18). Cardiac death and nonfatal
myocardial infarction (MI) were significantly higher with drug-eluting stents
(DES), which contrasts with a trend toward a lower restenosis-related target
vessel revascularization (TVR) rate after implantation of DES.
For the entire 18-month period, the rate of cardiac death/nonfatal MI did not differ between DES-treated and BMS-treated subjects (Figure, A). After excluding intervention-associated events during the first 30 days, death and MI were shown to occur more frequently among subjects with DES in the period from 1 to 18 months after stent implantation (Figure, C). It was evident that there was a continuous increase in cumulative events over the entire observation period in DES-treated subjects. By contrast, although the events curve gradually flattened out in BMS-treated subjects, there was no case of death or nonfatal MI after 1 year. Drug-eluting stent implantation was associated with a significantly elevated risk of death/nonfatal MI in the multivariate analysis (hazard ratio = 2.2; 95% confidence interval, 1.1-4.7; .03). On the other hand, the rate of TVR remained reduced after DES implantation (Figure, B, D).
Figure. Cardiac death/myocardial infarction (MI) and restenosis-related target vessel
revascularizations (TVR) after drug-eluting (DES) versus bare-metal stent (BMS) implantation.
Comparison of the occurrence of cardiac death/nonfatal myocardial infarction (MI) (A and C) and
the need for "restenosis-related" target vessel revascularization (TVR; B and D) after DES (red)
versus BMS (blue) implantation. Note that in this graph, the initial 30-day events that are not
related to drug-eluting properties of the stents are included (A and B; period 0 to 18 months) or
disregarded (C and D; period 1 to 18 months). (Reprinted from Pfisterer M, Brunner-La Rocca HP,
Buser PT, et al, for the BASKET-LATE Investigators. Late clinical events after clopidogrel
discontinuation may limit the benefit of drug-eluting stents: an observational study of drug-
J Am Coll Cardiol.
eluting versus bare-metal stents. 2006;48:2584-2591 with permission
from American College of Cardiology Foundation).
Sixteen of 65 clinical events (25%) were associated with stent thrombosis following clopidogrel withdrawal: 2 angiographically proven subtotal thrombotic obstructions with increasing angina but without MI, 11 nonfatal MIs, and 3 cardiac deaths. The rate for subjects with DES implantation was 2 to 3 times higher than for those with BMS implantation (1.4% vs 0.8% for angiographically documented stent thrombosis and 2.4% vs 0.8% for thrombosis-related clinical events, respectively). However, the difference was not statistically significant because of the low incidence. Late thrombotic events were observed over the entire follow-up period. They occurred 15 to 362 days after discontinuation of clopidogrel, with an average of 115 days.
The BASKET-LATE trial showed for the first time in a prospective randomized study that the incidence of cardiac death/MI after discontinuation of clopidogrel was higher among subjects with DES implantation than among those with BMS implantation. Late clinical events occurred during the entire follow-up period after withdrawal of clopidogrel (after 15-362 days). Compared with BMS-treated subjects, there was a continuous increase in the cumulative event rate over the entire observation period following DES implantation.
The occurrence of stent thromboses in DES subjects more than 1 year after implantation was initially described in clinical case reports.5,8 The importance of this phenomenon lies especially in the high mortality rate, which may be up to 45%.4 More recent studies investigated the pathophysiological background of late stent thrombosis after DES implantation. In autopsy studies, Joner and colleagues demonstrated that endothelialization is delayed after DES implantation compared with BMS.11 The least endothelialization was observed in DES-treated subjects with late stent thrombosis. Kotani and colleagues were able to directly visualize the delayed intimal healing in an angioscopic study.12 DES-treated subjects had less endothelialization and a more frequent occurrence of thrombi than BMS-treated subjects 3 to 6 months after percutaneous coronary intervention. Compared with 2 recent cohort studies,4,8 the BASKET-LATE study showed a higher incidence of late stent thromboses after DES implantation. However, in the BASKET-LATE study, more complex lesions were treated, 60% of the subjects had an acute coronary syndrome (one third of them with ST-segment elevation MI), and clopidogrel was withdrawn after 6 months in all subjects. Some of these high-risk characteristics in the BASKET-LATE population could be identified as predictors of late thrombotic events, among these the need for glycoprotein IIb/IIIa inhibitors reflecting acute coronary disease and/or sub-optimal stenting results (including stenting of bifurcations or bypass grafts).
The results of the BASKET-LATE study indicate that there may be a penalty for the lower rate of stenosis and reinterventions after DES implantation, namely, an increased rate of late stent thromboses. The implantation of DES in 100 patients may possibly prevent 5 major cardiac events after 6 months, but may lead to 3 cases of death or nonfatal MI during months 7 to 18. In light of this, the question arises whether longer administration of clopidogrel might reduce the number of late clinical events. The results of a recent observation study indicate that longer clopidogrel administration could possibly lead to a lower risk of death and MI.13 Because the thrombotic events in BASKET-LATE occurred in a broad time window after withdrawal of clopidogrel, it is not certain that the withdrawal of clopidogrel therapy was directly related to these clinical events. Moreover, the increased risk of hemorrhage and the high cost of prolonged dual-antiplatelet therapy must be considered. In addition to new treatment regimens for antiplatelet therapy, future strategies for the prevention of late stent thromboses should thus include further developments in stent technology, for example, bioabsorbable stents or modifications in the dose and release kinetics of stent medications.
The results of the BASKET-LATE study showed that the reduction in TVR after discontinuation of clopidogrel therapy among DES-treated subjects must be weighed against an increase in the incidence of late nonfatal MI and cardiac death, which may be related to late stent stenosis.