In randomized trials, longer drug-eluting stent (DES) length has been associated with adverse clinical events. We used data from the EVENT registry to examine the impact of DES length on outcomes in routine clinical practice. We identified 5,425 unselected consecutive patients from the EVENT registry who had a single vessel treated with DES for nonemergency indications from 2004 through 2007. The association between stented length and short- and long-term outcomes was analyzed in ordinal categories (<15, 15 to 19, 20 to 24, and >24 mm) and as a continuous variable. There were few differences in baseline characteristics across categories. At 1 year, there was a stepwise increase in major adverse cardiac events (composite of death, myocardial infarction [MI], and target lesion revascularization [TLR]) with increasing stent length (8.0%, 10.1%, 11.8%, and 14.8%, p <0.001) and a similar relation with TLR (3.0%, 3.1%, 3.3%, and 5.0%, p = 0.02). After adjusting for demographic, clinical, angiographic, and treatment characteristics, longer stent length remained associated with 1-year major adverse cardiac events (adjusted hazard ratio 1.17 per 10-mm increase stent length) and TLR (hazard ratio 1.20 per 10 mm), but not with stent thrombosis. In conclusion, longer DES length is associated with increased adverse events, predominantly periprocedural MI, but also an increased rate of TLR.
Among the most common off-label uses of drug-eluting stents (DESs) is their placement to treat long lesions or diffuse coronary artery disease. In pivotal DES approval trials, longer stent length was associated with an increased rate of adverse clinical events. However, because these trials employed strict inclusion/exclusion criteria, limits to stented segment length, and routine angiographic follow-up, the specific impact of DES length in routine clinical practice is not well established. We therefore sought to examine the specific clinical consequences of DES implantation for the treatment of long lesions in routine clinical practice and to determine whether the increased risk of additional stent length as seen with bare-metal stents (BMSs) is relevant in the current DES era.
Methods
The population for this study was derived from the Evaluation of Drug-Eluting Stents and Ischemic Events (EVENT) registry, which enrolled consecutive patients undergoing percutaneous coronary intervention (PCI) with intended stent implantation at >50 United States centers from 2004 through 2007. Specific details regarding the design and methodology of the EVENT trial have been reported previously.
In brief, patients >18 years of age and undergoing PCI with attempted implantation of any Food and Drug Administration–approved stent were eligible and enrolled at participating centers by 1 of 3 enrollment strategies designed to ensure an unselected patient population. Exclusion criteria included PCI or cardiac surgery within 4 weeks, enrollment in another device/drug study that excluded concurrent enrollment or involved blinded placement of DES, previous enrollment in EVENT, or failure/inability/unwillingness to provide informed consent. For this analysis, patients treated for emergency indications, with multivessel PCI, and with BMSs were also excluded. Emergency indications were excluded because of inability to distinguish periprocedural myocardial infarction (MI) owing to procedural complications from the consequences of ongoing MI at the time of the procedure.
Sociodemographic, clinical, and treatment characteristics were collected prospectively for each patient using standard case-report forms, which were then submitted to the data coordinating center for cleaning and analysis. Enzymatic data related to myonecrosis (creatine kinase-MB) were collected at baseline and periprocedurally every 8 hours for a minimum of 2 samples after the index procedure. These samples were assayed at the clinical laboratory of each site using local reference values.
Periprocedural MI was defined as an increase of creatine kinase-MB or creatine kinase ≥3 times the local upper limit of normal with preference given to creatine kinase-MB values or persistent ST-segment elevation >1 mm in 2 contiguous electrocardiographic limb leads or 2 mm in 2 contiguous precordial leads. In patients who had an increased creatine kinase or creatine kinase-MB at presentation, a diagnosis of MI required at least a twofold increase in enzyme levels after the procedure. Stent thrombosis was defined as definite or probable stent thrombosis according to the Academic Research Consortium definitions. All events were adjudicated by 2 independent cardiologists who were blinded to patient characteristics.
All patients were contacted by telephone at 6 and 12 months after the index PCI to ascertain the occurrence of specific clinical events including death, MI, revascularization, stent thrombosis, and noncardiac surgery. Continued use of aspirin and/or clopidogrel was also assessed at each follow-up time point. All participating sites obtained institutional review board approval, and all patients provided written informed consent.
The primary study end point was major adverse cardiovascular events at 1 year, which was defined as a composite of death, MI, or target lesion revascularization (TLR). Secondary end points included stent thrombosis and a composite of in-hospital death, MI, and TLR and individual components of the primary end point. In addition, given previous data demonstrating the long-term clinical importance of higher creatine kinase or creatine kinase-MB increases, an additional composite end point of death and “large MI” (defined as creatine kinase-MB ≥5 times local upper limit of normal) was analyzed.
Stent length was considered as a surrogate for lesion length and was calculated as the sum of lengths for all implanted stents (ignoring any potential overlap). Lesion length was not collected directly because such measurements were judged impractical in routine practice without an angiographic core laboratory and operator-determined lesion lengths have not been validated. Continuous variables are described as mean ± SD and were compared by a trend test through linear regression, and categorical variables are presented as percentage and were compared by Cochran-Mantel-Haenszel chi-square test with row mean scores. The association between stent length and clinical outcomes was analyzed in ordinal categories (<15, 15 to 19, 20 to 24, and >24 mm) and as a continuous variable. In sensitivity analysis we looked at the composite all-cause death, large MI (creatine kinase-MB >5 times upper limit of normal) before hospital discharge, or any MI postdischarge.
Time-to-event curves were constructed using the Kaplan–Meier approach. Cox proportional hazard models were fit to relate stent length to the outcomes of interest, adjusting for prespecified demographic, clinical, angiographic and treatment covariates (saturated models). Demographic and clinical covariates included age, gender, diabetes, previous MI, previous coronary bypass surgery, acute coronary syndrome indication, congestive heart failure, creatinine clearance, and long-term dialysis treatment. Angiographic covariates used for adjustment included left anterior descending coronary artery target vessel, saphenous vein target vessel, number of diseased vessels, sirolimus-eluting stent (Cypher, Cordis Corporation, Miami, Florida) versus paclitaxel-eluting stent (Taxus, Boston Scientific Corp, Natick, Massachusetts) use, and maximum balloon diameter. Treatment covariates examined were planned use of glycoprotein IIb/IIIa receptor inhibitor and planned use of bivalirudin and clopidogrel loading preprocedure.
All statistical analyses were performed using SAS 9.1 (SAS Institute, Cary, North Carolina). Statistical significance was defined as a 2-tailed p value ≤0.05, and no adjustment was performed for multiple comparisons in this exploratory study.
Results
The overall EVENT population consisted of 10,148 patients. For this analysis, we excluded 3,076 patients who underwent multilesion revascularization, 608 patients who were treated for ST-segment elevation MI, and 1,039 patients who were treated with BMSs. After these exclusions, the study population consisted of 5,425 patients who underwent nonemergency single-vessel PCI with ≥1 DES ( Figure 1 ) . Of these, total stented lengths were <15 mm in 1,480, 15 to 19 mm in 1,420, 20 to 24 mm in 1,048, and >24 mm in 1,477. Complete 1-year follow-up was available in 5,122 (94.4%) of the study population.
Baseline characteristics of the study population stratified by total stented length are presented in Table 1 . There were few significant differences in clinical, angiographic, or procedural variables among categories. At 1 year there was a strong association between increasing stent length and the primary composite end point of death, MI, or TLR (major adverse cardiovascular events) in univariate analyses ( Table 2 ). Examination of individual end-point components ( Table 2 ) demonstrated a similar association with MI and TLR, but there was no association with death or stent thrombosis. Inspection of timing of events demonstrated that early differences were driven predominantly by differences in periprocedural MI, whereas later differences were related mainly to TLR ( Figure 2 ) .
| Variable | Total Stent Length (mm) | Overall Population (n = 5,425) | p Value | |||
|---|---|---|---|---|---|---|
| <15 (n = 1,480) | 15–19 (n = 1,420) | 20–24 (n = 1,048) | >24 (n = 1,477) | |||
| Age (years) | 64.9 ± 11.4 | 64.2 ± 11.3 | 64.5 ± 11.3 | 63.9 ± 11.3 | 64.4 ± 11.3 | 0.06 |
| Men | 64.1% | 67.9% | 68.8% | 70.3% | 67.7% | <0.01 |
| Diabetes mellitus | 36.3% | 33.9% | 35.8% | 33.5% | 34.8% | 0.31 |
| Congestive heart failure | 11.3% | 7.7% | 10.8% | 9.1% | 9.7% | <0.01 |
| Previous myocardial infarction | 35.2% | 33.3% | 33.4% | 34.7% | 34.2% | 0.67 |
| Previous coronary artery bypass grafting | 22.5% | 21.9% | 23.1% | 21.0% | 22.1% | 0.64 |
| Creatinine clearance (ml/min) | 86.9 ± 49.7 | 90.4 ± 46.7 | 89.1 ± 65.6 | 90.4 ± 38.7 | 89.2 ± 49.9 | 0.13 |
| Dialysis | 2.5% | 1.7% | 2.2% | 0.9% | 1.8% | <0.01 |
| Acute coronary syndrome | 36.8% | 36.2% | 36.4% | 33.4% | 35.6% | 0.21 |
| Left anterior descending coronary artery lesion | 40.0% | 37.0% | 37.8% | 37.6% | 38.1% | 0.37 |
| Saphenous vein graft lesion | 5.8% | 8.0% | 8.6% | 7.9% | 7.5% | 0.03 |
| Number of diseased vessels | 1.5 ± 0.8 | 1.5 ± 0.8 | 1.6 ± 0.8 | 1.6 ± 0.8 | 1.6 ± 0.8 | 0.11 |
| Maximum balloon diameter (mm) | 3.02 ± 0.57 | 3.09 ± 0.52 | 3.06 ± 0.51 | 3.12 ± 0.50 | 3.07 ± 0.53 | <0.01 |
| Drug-eluting stent type = paclitaxel-eluting stent | 40.1% | 41.3% | 60.2% | 35.2% | 43.0% | <0.01 |
| Number of stents implanted | 1.0 ± 0.03 | 1.01 ± 0.10 | 1.09 ± 0.34 | 1.71 ± 0.73 | 1.21 ± 0.51 | <0.01 |
| Clopidogrel loading before procedure | 59.9% | 58.6% | 58.6% | 59.6% | 59.2% | 0.86 |
| Planned glycoprotein IIb/IIIa inhibitor | 18.8% | 21.3% | 22.6% | 24.0% | 21.6% | 0.01 |
| Bivalirudin as planned anticoagulant | 36.8% | 34.6% | 33.4% | 34.3% | 34.9% | 0.29 |
| Variable | Total Stent Length (mm) | p Value | |||
|---|---|---|---|---|---|
| <15 | 15–19 | 20–24 | >24 | ||
| (n = 1,480) | (n = 1,420) | (n = 1,048) | (n = 1,477) | ||
| Major adverse cardiac events | 8.0% | 10.1% | 11.8% | 14.8% | <0.01 |
| Death | 2.0% | 2.7% | 3.0% | 2.2% | 0.41 |
| Myocardial infarction | 3.9% | 6.1% | 6.5% | 9.7% | <0.01 |
| Target lesion revascularization | 3.0% | 3.1% | 3.3% | 5.0% | 0.02 |
| Stent thrombosis | 0.4% | 0.9% | 0.2% | 0.6% | 0.1 |
| Large periprocedural myocardial infarction or late spontaneous myocardial infarction | 2.5% | 4.3% | 4.7% | 6.2% | <0.01 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
