Journal of the American College of Cardiology Vol. 35, No. 3, by the American College of Cardiology ISSN /00/$20.

Transcription

1 Journal of the American College of Cardiology Vol. 35, No. 3, by the American College of Cardiology ISSN /00/$20.00 Published by Elsevier Science Inc. PII S (99)00580-X Procedural Results and Late Clinical Outcomes After Percutaneous Interventions Using ( 25 mm) Versus ( 20 mm) Stents Ran Kornowski, MD, FACC, Balram Bhargava, MD, DM, Shmuel Fuchs, MD, Alexandra J. Lansky, MD, FACC, Lowell F. Satler, MD, FACC, Augusto D. Pichard, MD, FACC, Mun K. Hong, MD, FACC, Kenneth M. Kent, MD, PhD, FACC, Roxana Mehran, MD, FACC, Gregg W. Stone, MD, FACC, Martin B. Leon, MD, FACC Washington, DC OBJECTIVES BACKGROUND METHODS To evaluate clinical outcomes after the use of long coronary stents. The use of long slotted-tube stents has been recently aroved in the U.S. to treat long lesions or dissections. Procedural success and long-term outcomes of long versus short stents have not been established. We evaluated rocedural success, major in-hosital comlications, target lesion revascularization and long-term (one year) clinical outcomes in 1,226 consecutive atients (1,259 native coronary lesions) who underwent a single vessel intervention using a single long ( 25 mm, 116 atients) or short ( 20 mm, 1,110 atients) tubular-slotted stent. RESULTS Patients treated with long stents had more diffuse ( 10 mm length) lesions (63% vs. 28%, 0.001). The mean stent length was 28 5 mm versus 15 2 mm for long versus short stent grous ( 0.001). Overall rocedural success was similar in the long versus short stent grous (96% vs. 98%, 0.08). However, major in-hosital comlications tended to occur more frequently in atients treated with longer stents (3.4% vs. 1.0%, 0.04). The rate of erirocedural non-q-wave myocardial infarction (MI) (creatine kinase-mb 5 times normal) was notably higher after long stent imlantation (23% vs. 11%, 0.001). Target lesion revascularization at one year was 14.5% vs. 13.8% ( 0.69), and target vessel revascularization rate was 19.6% vs. 17.3% ( 0.41) in the long versus short stent grou, resectively. There was no difference in one year mortality (2.5% vs. 3.5%, 0.49) or Q-wave MI (2.7% vs. 1.2%, 0.48), and the overall cardiac event-free survival was similar for the two grous (81%). CONCLUSIONS The use of single coronary long ( 25 mm) versus short ( 20 mm) stents is associated with: 1) somewhat increased major rocedural comlications, 2) significantly higher frequency of erirocedural non-q-wave MIs, and 3) equivalent reeat revascularization risk and cardiac event-free survival out-of-hosital u to one year. (J Am Coll Cardiol 2000;35:612 8) 2000 by the American College of Cardiology The treatment of atherosclerotic coronary laques with metallic slotted-tube stents has been shown to imrove the acute and long-term outcomes obtained by catheter-based coronary interventions (1 4). Imroved imlantation techniques using high-ressure balloon inflation and the addition of new antilatelet harmacotheray enable stent imlantation without long-term anticoagulation, with very low rates of erirocedural comlications (5 7). Initiatory trials have shown favorable stent results comared From the Cardiac Catheterization Laboratory and The Cardiovascular Research Foundation, The Washington Hosital Center, Washington, DC. This study was suorted by a grant from the Cardiology Research Foundation, The Washington Cardiology Center, Washington, DC. Manuscrit received Aril 21, 1999; revised manuscrit received August 16, 1999; acceted November 3, with balloon angiolasty in short lesions and relatively large sized vessels (1,2). It is still unclear whether these results can be generalized to longer lesions in diffuse coronary disease (8). Data on the effectiveness of stent treatment in diffuse disease scenarios are limited. Prior studies have reorted higher rocedural comlications and stent thrombosis rates and more frequent late restenosis in atients treated with multile stents laced in diffuse lesions (9 13). Recently, we showed that atients with relatively large sized vessels treated with multile contiguous stents have favorable rocedural results and major cardiac event rates during follow-u (14). Accurate ositioning of multile stents may become difficult, however, with long overlaing stent segments or uncovered gas leading to inadequate results comared with

2 JACC Vol. 35, No. 3, 2000 March 1, 2000:612 8 Kornowski et al. Clinical Outcomes With Stents 613 Abbreviations and Acronyms CK creatinine kinase IVUS intravascular ultrasound MB myocardial band MI myocardial infarction OR odds ratio QCA quantitative coronary angiograhy TIMI thrombolysis in myocardial infarction TLR target lesion revascularization TVR target vessel revascularization single stent imlantation in most otherwise reorted series. Recently, with the availability of long stents in the U.S., we have comared the acute rocedural and long-term (one year) outcomes of atients with a single long ( 25 mm) versus short ( 20 mm) slotted-tube stent imlantation in native coronary arteries. METHODS Patients and follow-u. The atient cohort includes a consecutive series of 1,226 atients (1,259 native coronary lesions) in the Cardiology Research Foundation Angiolasty Database treated with a single stent imlanted in a single native coronary artery between July 1, 1997, and July 1, Patients were divided into two grous: long stents ( 25 mm) (n 116 atients) and a short ( 20 mm) stent grou (n 1,110 atients). All indications for stent use (elective use to otimize angiograhic results and reduce late restenosis, rovisional use to treat subotimal rimary device result or urgent use to treat abrut or threatened vessel closure) are included in this study. Baseline clinical demograhics and in-hosital comlications were confirmed by indeendent hosital chart review. All atients underwent re- and ost-intervention, 12- lead electrocardiogram to detect rocedural related ischemic changes or the aearance of a new athologic Q-wave on the surface ECG. Blood samles were routinely acquired from all atients after the rocedure for creatine kinase (CK)-MB enzyme at 8, 16 and 24 h (normal values 0 to 4 ng/ml). The diagnosis of non-q-wave myocardial infarction (MI) was based on CK-MB elevation 5 times normal values in the absence of new athologic Q-waves on ost-intervention electrocardiograms. Perirocedural CK-MB elevation is also reorted as three times normal cut-off. Out-of-hosital clinical outcomes u to eight months were obtained by serial telehone interviews by research nurses and late clinical events (death, Q-wave MI), target lesion revascularization (TLR), target vessel revascularization (TVR) or any cardiac event (death, Q-wave MI, TLR) was adjudicated and corroborated by accomanying source documentation. Stent techniques. After the initial balloon angiolasty or ablative rocedure, coronary stents were imlanted over in. extra-suort guidewire. All stents used during the study eriod were included in the current analysis. Adjunct balloon inflation (usually 16 atmosheres) was added after initial stent deloyment in all cases. Otimal stent imlantation was carefully monitored using an iterative technique with intravascular ultrasound (IVUS) monitoring in the majority of cases. The re- and ost-stent anticoagulation regimens included asirin (325 mg daily) and ticloidine (250 mg twice daily) for one month. Patients with intervention in more than one vessel or more than a single stent were excluded from analysis. Quantitative angiograhic analysis. Five hundred ninetysix lesions were available for comlete quantitative and qualitative angiograhic analysis. Standard morhologic criteria were used for the identification of lesion location, length, eccentricity, calcification and ulceration. Quantitative angiograhic analysis was erformed using selected end-diastolic frames demonstrating the stenosis in its most severe rojection. Using the contrast-filled guiding catheter as the calibration standard, roximal and distal references (within 5 mm of lesion margins) were measured and lesion minimal lumen diameters and ercent diameter stenosis were determined before and after intervention. Statistics. Continuous variables are resented as mean 1 standard deviation. Categorical data are resented as ercent frequency and comared between grous using chisquare statistics. Survival curves were calculated and dislayed using the SAS LIFETEST (SAS Institute, Cary, North Carolina) rocedure and Log-Rank statistics were used for testing of survival homogeneity between the two grous. Wilcoxon statistics were used for survival comarison between grous (long versus short stents). Mean values were comared using the unaired Student t test. A value 0.05 was acceted as statistically significant. RESULTS Baseline demograhics. Table 1 lists the baseline characteristics of all treated atients, distinguished according to long versus short stents. Patients treated with long stents on the average were younger and suffered more often from hyertension and revious MI but had sustained fewer rior coronary angiolasty rocedures. Otherwise, the demograhics were similar between these two grous. Procedural data. Before stent deloyment, atheroablation was erformed in 17% and 16% of the atients in the two grous (Table 2). Overall, the tyes of stents used were different between grous with the majority (80%) of atients in the short stent grou treated with the Palmaz-Schatz stent (Cordis Cor., Warren, New Jersey) and the long stent grou treated with either the multilink (33%), Micro-II/ GFX (Alied Vascular Engineering Inc., Santa Rosa, California) (32%) or the NIR stent (13%). The mean stent

3 614 Kornowski et al. JACC Vol. 35, No. 3, 2000 Clinical Outcomes With Stents March 1, 2000:612 8 Table 1. Baseline Characteristics of the Study Poulation Vessels Treated (Patients) (N 116) (N 1110) Mean age (yrs) Male gender (%) Unstable angina (%) Hyertension (%) Diabetes mellitus (%) Hyercholesterolemia (%) Prior MI (%) Prior CABG (%) Prior angiolasty (%) LVEF (%) CABG coronary artery byass grafting; LVEF left ventricular ejection fraction; MI myocardial infarction. length (long vs. short) was 28 5 mm versus 15 2mm ( 0.001). The number of rovisional/lanned versus urgent stenting was similar in the two grous (97% vs. 96% and 3% vs. 4%; NS). Imortantly, the erirocedural use of abciximab was significantly more frequent in atients treated with long versus short stents (14 vs. 3.9%, 0.001). Lesion characteristics. Table 3 lists the lesion location data for all 1,259 native coronary lesions and full qualitative and quantitative measurements obtained in 535 lesions. The lesions were similarly distributed among the coronary arteries between grous but short stents were more frequently imlanted in the ostial location. er stents were more often imlanted in longer ( 10 mm) lesions or in those with rerocedural Thrombolysis in Myocardial Infarction (TIMI) 0/1 flow but less often in restenotic or calcified lesions. Angiograhic comlications (dissections, abrut closure or no reflow) were similarly observed in both grous. By quantitative angiograhy, the average roximal reference diameter was similar between grous, but the distal reference was smaller for the long stent grou ( vs mm, 0.01) as the vessel taered to a longer distance in this grou. Postrocedural lesion measurements were different; the final oststent minimal lumen diameter was smaller ( mm vs mm, 0.001), and accordingly, the final stent diameter stenosis was higher (19 13% vs. 9 16%, 0.001), signifying more residual stenosis immediately after longer stenting (Table 3). Procedural results. Overall rocedural success was similar in the long versus short stent grous (96% vs. 98%, 0.08) (Table 4). However, major in-hosital comlications tended to occur more frequently in atients treated with longer stents (3.4% vs. 1.0%, 0.04). The rate of erirocedural non-q-wave MI (CK-MB 5 times normal) was notably higher after longer stent imlantation (23% vs. 11%, 0.001). The length of hositalization, however, was similar for both grous ( vs days for long vs. short stents, resectively, 0.08). The rates of in-hosital reeat target vessel angiolasty and stent thrombosis were similar for the two grous. -term outcomes. Clinical follow-u at one year was available in 102 of 116 atients (88%) treated with long stents and in 1,024 of 1,110 atients (92%) treated with shorter stents. During follow-u, there was no difference in mortality between grous (2.5% for long versus 3.5% for short stents, 0.49) (Table 4). The rate of Q-wave MI was also similar for long stent grou versus short stenting (2.7% vs. 1.2%, 0.18). Overall TLR at one year was 14.5% for long stents versus 13.8% in short stent grou ( 0.69, Fig. 1A). Similarly, TVR was higher but similar for both grous ( 0.41). Likewise, actuarial event-free survival curves for any event during one-year follow-u (death, Q-wave MI, TLR) was similar for both grou (81.3% for long stents versus 81.5% for short stents, 0.44, Fig. 1B). Multivariate analysis. Logistic regression analysis was used to identify indeendent redictors of any cardiac event Table 2. Interventional Procedures Vessels (Lesions) (N 117) (N 1142) Procedure Tye (re-stent) Balloon alone (%) Atheroablation (%) IVUS erformed (%) Tye of Stent Palmaz-Schatz/Crown (%) Multilink-Duette (%) NIR (%) Micro-II/GFX (%) Wallstent (%) Others (%) Mean Stent Length (mm) Abciximab Used (%) IVUS intravascular ultrasound.

4 JACC Vol. 35, No. 3, 2000 March 1, 2000:612 8 Kornowski et al. Clinical Outcomes With Stents 615 Table 3. Qualitative and Quantitative Characteristics of Stented Lesions Vessels (Lesions) (N 117) (N 1142) Target Vessel RCA (%) LAD (%) LCx (%) Lt Main Ostial (%) Proximal (%) Lesion Characteristics (N 92) (N 504) Restenotic (%) Calcium (%) Length (mm) Length 10 mm (%) Ulceration (%) Eccentricity (%) Tye B 2 /C (%) TIMI 0 or Procedural Comlications Dissection tye C (%) Abrut closure (%) No reflow (%) Quantitative Measurements Proximal reference diameter (mm) Distal reference diameter (mm) Prerocedure (mm) Final oststent (mm) Lesion % diameter stenosis Prerocedure (%) Final oststent (%) LCx left circumflex; Lt Main left main; RCA right coronary artery; TIMI thrombolysis in myocardial infarction. (death, Q-wave MI, TLR) or TLR alone following a single vessel/lesion stenting among the treated atients (Table 5). Variables included in the model were unstable angina, age, gender, history of angiolasty, diabetes mellitus, use of abciximab, roximal reference vessel diameter, distal reference vessel diameter, final ercent diameter stenosis, stent length, stent tye, ostial location, and lesion length. Diabetes (odds ratio [OR] 1.9), history of angiolasty (1.9), roximal reference diameter (0.68) and unstable angina (OR 1.6) were indeendent redictors of any adverse cardiac event during followu. History of angiolasty (OR 2.1) and roximal reference vessel diameter (OR 0.57) were the only redictors for TLR. Neither stent length nor lesion length redicted subsequent adverse cardiac events from the time of hosital discharge to one-year follow-u. Stent to artery ratio. Stent to lesion length ratio (as determined by the known stent length and the measured lesion length by QCA) was lotted against TLR and TVR at follow-u. According to this analysis, there were no significant differences in rates of TLR or TVR between grous distinguished on the basis of stent to lesion length ratio, (Fig. 2). DISCUSSION The treatment of diffuse coronary artery disease has been traditionally associated with disaointing acute and longterm results in most reorted conventional balloon angiolasty and new-devices series, with increased risk for acute comlications and restenosis (15 18). In the stent era, two main aroaches to stenting of diffuse lesions have been advocated: sot stenting of only severe stenosis or full lesion coverage with the goal of anchoring stents into normal reference segments (19). Until long stents became available, full coverage of diffuse lesions or long dissection could have been achieved only by the use of multile overlaing stents (9 13) or long self-exandable stents (20,21). Recent exeriences suggested that such strategy advocated to treat long lesions is feasible, with relatively low rocedural comlications but with relatively high restenosis rate, unless used to scaffold diffuse lesions in large ( 3.25 mm) sized vessels (14). The various roblems that were observed in other studies with multile stenting in the same vessel included increased risk of subacute thrombosis (11), restenosis (12,13) and technical difficulty in deloying multile stents with increased catheterization time, dye

5 616 Kornowski et al. JACC Vol. 35, No. 3, 2000 Clinical Outcomes With Stents March 1, 2000:612 8 Table 4. In-Hosital Procedural Results and Clinical Outcomes at 12 Months Follow-u Vessels (Patients) (N 116) (N 1,110) In-Hosital Procedural success (%) Major Hosital Comlications Death (%) Q-wave infarction (%) Emergent coronary byass (%) Combined (%) Non Q-wave MI (%) CK-MB 3 normal (%) Stent thrombosis (%) Reeat angioasty (%) One-Year Follow-u Death (%) Q-wave infarction (%) TLR (%) TVR (%) Cardiac event-free survival (%) CK creatine kinase; TLR target lesion revascularization; TVR target vessel revascularization. volume and radiation exosure (22). Moreover, the greater metal density and reeated stent on stent trauma with overlaing stents was roosed to imair vessel wall integrity causing a higher degree of vascular injury and romoting more neointimal roliferation (23,24). The availability of long stents in the U.S. with imroved flexibility, trackablity and scaffolding roerties has rovided us with the oortunity to comare the acute rocedural and long-term results following treatment of coronary narrowing using longer slotted-tube stents in native coronary arteries. Several studies have imlicated stent length or the number of stents imlanted as contributing factors to restenosis in addition to intrinsic lesion characteristics that redisose neointimal formation (23 25). Our study s main finding is that, unlike revious reorts, the use of a single coronary long ( 25 mm) versus short ( 20 mm) stent is associated with: 1) somewhat increased major rocedural comlications, 2) significantly higher frequency of erirocedural non-q-wave MIs or any CK-MB elevation, and 3) equivalent reeat revascularization risk and event-free survival out-of-hosital u to one year. According to our exerience, the erirocedural use of abciximab was significantly more revalent in atients treated with long versus short stents. This may artially account for imroved Figure 1. (A) Actuarial event-free survival curves for target lesion revascularization (TLR), (B) or any adverse event (death, Q-wave MI or TLR) for one year following long ( 25 mm) versus short ( 20 mm) slotted-tube single vessel stenting. Table 5. Indeendent Predictors of Any Cardiac Event and TLR During One-Year Follow-u Period Predictive Variables Odds Ratio 95% Confidence Limits Any Cardiac Event Diabetes mellitus History of angiolasty Proximal reference vessel diameter Unstable angina TLR History of angiolasty Proximal reference vessel diameter

6 JACC Vol. 35, No. 3, 2000 March 1, 2000:612 8 Kornowski et al. Clinical Outcomes With Stents 617 Figure 2. Stent-to-lesion length ratio lotted against target lesion (TLR) and target vessel (TVR) revascularization rates. rocedural and long-term results obtained by longer stents in our atient cohort. However, the overall use of abciximab has been relatively low in both grous at the time of this study. Imortantly, the stent to lesion length ratio has not been shown to be a major determinant of subsequent TLR or TVR in our exerience. It is also notable that the roximal reference vessel diameter (but not the distal one) was among the strongest redictors of TLR according to a multivariate model, imlying that the vessel size in its roximal lesion inlet lays an imortant role in achieving favorable results in diffuse disease scenarios requiring long stents. Interestingly, stent length was not redictive of TLR in our model. This finding is in accordance with our revious observation in single and multivessel stenting (14,26). In this study, there was a notably higher frequency (23%) of rocedural related non-q-wave MI in atients treated with long stents. Such higher revalence of erirocedural CK-MB elevation may reflect a more comlex clinical and anatomic milieu in atients treated with longer lesions such as diffuse disease, thrombus containing lesions, more extensive dissections, chronic total occlusions and side branch vessel occlusion. In this resect, it was recently found that re-intervention lesion characteristics and esecially the measured amount of laque at the lesion site are the major determinants of CK-MB elevation after coronary stenting (27). Desite other reorts indicating a strong association between erirocedural CK-MB elevations and late adverse cardiac events (28 30), thus far, our reliminary out-ofhosital exerience with long stents did not indicate higher mortality or Q-wave MI rates comared with shorter stents. However, it should be noted that if major CK-MB elevation had to be accounted for by major adverse cardiac events, the overall event rates would have been much higher in the long stent grou in our study. Our results corroborate with revious reorts by showing that stenting of long lesions using the new generation long stents can be accomlished with high rocedural success and relatively low comlications rates. Rozenman et al. (31) have treated 57 atients with 67 long ( 30 mm) various tyes of stents. Procedural success has been obtained in all but one atient. Beyar et al. (32) have reorted the results obtained by using the long version of the BeStent (Medtronic Inc., Minneaolis, Minnesota) to treat diffuse lesions. Procedural success was high and achieved in 97% of atients, but overall restenosis rate was higher in atients with longer lesions who were treated with longer (mostly 25 mm) stents. Kobayashi et al., (33) using a 32 mm NIR stent have reorted angiograhic restenosis rates of 51% comared with 13.3% in atients treated with 16 mm NIR stents. Nevertheless, those atients also differed in reference vessel size, with bigger reference diameter found in the shorter stent grou, robably accounting for the difference in restenosis rate. More recently, LeBreton et al. (34) have reorted their exerience with the use of long (32 mm) NIR stents in 187 atients enrolled in a large multicenter French registry. Angiograhic success has been obtained in 99% of treated lesions. At follow-u (mean seven months), stent thrombosis occurred in 0.5%, death and MI occurred in 3% and 0.6%, resectively, and TLR rate was only 6%. Overall, cardiac event-free survival has been 86% and very similar to our own exerience. Finally, a nonrandomized registry comaring the long (39 mm) AVE Micro II stent with the Palmaz-Schatz (15 mm) stent found similar rocedural results and a similar TLR rate (9.2% vs. 8.1%, NS) desite longer lesions on average (23 vs. 12 mm) and more high risk baseline demograhics among atients treated with the longer stent (35). Two IVUS studies from our laboratory may exlain the observed findings in which clinical restenosis rates have not been excessive with longer stent imlantation or higher stent-to-lesion length coverage ratio. Hong et al. (36) have measured in vivo stent length by IVUS and found that when stent length has been adjusted to IVUS measured ostintervention lumen area, the increase in length did not imact TLR. Moreover, it was found that increasing the stent to lesion length ratio while otimizing full lesion coverage actually decreased the TLR rate (37). Those observations hel exlain our study findings in which relatively low TLR rates were observed among atients treated with single long stents, and they may indicate the full lesion coverage aroach for diffuse lesions. Study limitations. Since this study was a retrosective analysis, it is unknown whether the use of a different theraeutic strategy for diffuse disease (e.g., sots stenting, more use of atheroablation devices before stenting) would result in comarable rocedural or late outcomes obtained by long stenting. Moreover, the significantly increased use of abciximab with longer stents is another imortant factor that may have had a beneficial imact on the results obtained by the use of longer stents. However, the overall use of abciximab in this study was relatively low. Imortantly, the comarison of long versus short stents is necessarily confounded by significant differences in lesion characteristics between the two grous, with the short stent grou having more ostial, restenotic and calified lesions but

7 618 Kornowski et al. JACC Vol. 35, No. 3, 2000 Clinical Outcomes With Stents March 1, 2000:612 8 less total occlusions and rior MI. It is also ossible that, if long stents were to be imlanted in smaller vessels, the acute and esecially long-term clinical outcomes would be less favorable comared with short stents. Also, it is ossible that the higher incidence of total occlusions with TIMI 0/1 flow in the long stent grou might have diminished the caacity for revascularization in the event that restenosis or reocclusion occurred, masking additional restenosis in the longer stent grou. The lack of angiograhic follow-u with an accurate assessment of restenosis is another otential limitation of our study. This is also the reason we could not assess retenosis atterns (focal vs. diffuse) in each grou. In addition, the ower of the multivariate analysis was limited by the relatively small number of lesions with quantitative angiograhic assessment comared with the clinical data. Finally, our study did not comare different stent designs to determine which may have roerties better suited for the treatment of long lesions. Rerint requests and corresondence: Dr. Ran Kornowski, Cardiovascular Research Foundation, Washington Hosital Center, 110 Irving Street, Northwest, Suite 4B-1, Washington, DC rxk3@mhg.edu. REFERENCES 1. Fischman DL, Leon MB, Baim DS, et al., for the STent REStenosis Study Investigators. A randomized comarison of coronary-stent lacement and balloon angiolasty in the treatment of coronary artery disease. N Engl J Med 1994;331: Serruys PW, de Jaegere P, Kiemeneij F, et al., for the BENESTENT Study Grou. A comarison of balloon-exandable-stent imlantation with balloon angiolasty in atients with coronary artery disease. N Engl J Med 1994;331: Versaci F, Gasardone A, Tomai F, et al. A comarison of coronary artery stenting with angiolasty for isolated stenosis of the roximal left anterior descending coronary artery. N Engl J Med 1997;336: Erbel R, Haude M, Ho HW, et al. Coronary-artery stenting comared with balloon angiolasty for restenosis after initial balloon angiolasty. Restenosis Stent Study Grou. N Engl J Med 1998;339: Colombo A, Hall P, Nakamura S, et al. Intracoronary stenting without anticoagulation accomlished with intravascular ultrasound guidance. Circulation 1995;91: Schomig A, Neumann FJ, Kastrati A, et al. A randomized comarison of antilatelet and anticoagulant theray after the lacement of coronary-artery stents. N Engl J Med 1996;334: Leon MB, Baim DS, Poma JJ, et al. A clinical trial comaring three antithrombotic-drug regimens after coronary-artery stenting. Stent Anticoagulation Restenosis Study Investigators. N Engl J Med 1998; 339: Peine CJ, Holmes DR Jr. ACC exert consensus document. Coronary artery stents. J Am Coll Cardiol 1996;28: Ellis SG, Savage M, Fischman D, et al. Restenosis after lacement of Palmaz-Schatz stents in native coronary arteries. Circulation 1992;86: Eeckhout E, Stauffer JC, Vogt P, et al. Placement of multile and different stent tyes for very long dissections during coronary angiolasty. Cathet Cardiovasc Diagn 1996;39: Moussa I, Di Mario C, Reimers B, et al. Subacute stent thrombosis in the era of intravascular ultrasound-guided coronary stenting without anticoagulation: frequency, redictors and clinical outcome. J Am Coll Cardiol 1997;29: Liu MW, Luo JF, Dean LS, et al. -term follow-u study of coronary reconstruction with multile stents. Am Heart J 1999;137: Mathew V, Hasdai D, Holmes DR, et al. Clinical outcome of atients undergoing endoluminal coronary artery reconstruction with three or more stents. J Am Coll Cardiol 1997;30: Kornowski R, Mehran R, Hong MK, et al. Procedural results and late clinical outcomes after lacement of three or more stents in single coronary lesions. Circulation 1998;97: Ellis S, Roubin G, King S, Douglas J. Imortance of stenosis morhology in the estimation of restenosis after elective ercutaneous transluminal coronary angiolasty. Am J Cardiol 1989;63: Teirstein PS, Warth DC, Haq N, et al. High seed rotational atherectomy for atients with diffuse coronary artery disease. J Am Coll Cardiol 1991;18: Aelman YEA, Piek JJ, Strikwerda S, et al. Randomized trial of excimer laser angiolasty versus balloon angiolasty for treatment of obstructive coronary artery disease. Lancet 1996;347: Landau C, Lange RA, Hillis LD. Percutaneous transluminal coronary angiolasty. N Engl J Med 1994;330: Silva JA, White CJ. Full coverage in long diffuse or focal sequential lesions. Int J Cardiovasc Interv 198;1: Gambhir DS, Sudha R, Trehan V, et al. Immediate and six-month outcome of self-exanding Wallstent for long lesions in native coronary arteries. Indian Heart J 1997;49: Ozaki Y, Keane D, Ruygrok P, et al. Six-month clinical and angiograhic outcome of the new, less shortening Wallstent in native coronary arteries. Circulation 1996;93: De Scheerder IK, Wang K, Kostooulos K, et al. Treatment of long dissections by use of a single long or multile short stents: clinical and angiograhic follow-u. Am Heart J 1998;136: Kastrati A, Schomig A, Elezi S, et al. Predictive factors of restenosis after coronary stent lacement. J Am Coll Cardiol 1997;30: Bauters C, Hubert E, Prat A, et al. Predictors of restenosis after coronary stent imlantation. J Am Coll Cardiol 1998;31: Kasaoka S, Tobis JM, Akiyama T, et al. Angiograhic and intravascular ultrasound redictors of in-stent restenosis. J Am Coll Cardiol 1998;32: Kornowski R, Mehran R, Satler LF, et al. Procedural results and late clinical outcomes following multivessel coronary stenting. J Am Coll Cardiol 1999;33: Mehran R, Mintz GS, Hong MK, et al. Clinical, rocedural, angiograhic and intravascular ultrasound redictor (and sequelae) of CK-MB rise ( 5 normal) following catheter-based interventions. Circulation 1998;98:I Tardiff BE, Califf RM, Tcheng JE, et al. Clinical outcomes after detection of elevated cardiac enzymes in atients undergoing ercutaneous intervention. IMPACT-II Investigators. Integrilin (etifibatide) to Minimize Platelet Aggregation and Coronary Thrombosis-II. J Am Coll Cardiol 1999;33: Califf RM, Abdelmeguid AE, Kuntz RE, et al. Myonecrosis after revascularization rocedures. J Am Coll Cardiol 1998;31: Abdelmeguid AE, Tool EJ. The myth of the myocardial infarctlet during ercutaneous coronary revascularization rocedures. Circulation 1996;94: Rozenman Y, Mereuta A, Mosseri M, et al. Initial exerience with long coronary stents: the changing ractice of coronary angiolasty. Am Heart J 1997;134: Beyar R, Rougin A, Hamburger J, et al. er lesion coverage is associated with an increase in six months clinical events from a multicenter evaluation of the serentine baloon-exandable stent (bestent) [abstr]. Eur Heart J 1997;18 Sul: Kobayashi Y, De Gregorio J, Reimers B, et al. Immediate and follow-u results following imlantation of the long and short NIR stent: comarison with Palmaz-Schatz stents [abstract]. J Am Coll Cardiol 1998;31 Sul A:312A. 34. Le Breton H, Bedossa M, Commeau P, et al. Clinical and angiograhic results of stenting for long coronary arterial atherosclerotic lesions. Am J Cardiol 1998;82: Heuser R, Kuntz R, Lansky A, et al. A comarison of the long AVE Micro Stent II and the Palmaz-Scatz stent: a Smart Trial registry [abstr]. J Am Coll Cardiol 1998;31:80A. 36. Hong MK, Mintz GS, Satler LF, et al. In vivo intravascular ultrasound measured stent lengths do not redict subsequent target lesion revascularization [abstr]. J Am Coll Cardiol 1999;33:101A. 37. Hong MK, Mintz GS, Satler LF, et al. Full lesion coverage reduces subsequent target lesion revascularization: an intravascular ultrasound study [abstr]. J Am Coll Cardiol 1999;33:61A.