Final results of a randomized trial comparing the NIR stent to the Palmaz-Schatz stent for narrowings in native coronary arteries

Final Results of a Randomized Trial Comparing the NIR Stent to the PalmazSchatz Stent for Narrowings in Native Coronary Arteries Donald S. Baim, MD, Donald E. Cutlip, MD, Charles D. O’Shaughnessy, MD, James B. Hermiller, MD, Dean J. Kereiakes, MD, Alessandro Giambartolomei, MD, Stanley Katz, MD, Alexandra J. Lansky, MD, Michelle Fitzpatrick, RN, MS, Jeffrey J. Popma, MD, Kalon K.L. Ho, MD, MSc, Martin B. Leon, MD, and Richard E. Kuntz, MD, MSc, for the NIRVANA Investigators* The NIR stent is a novel second generation tubular stent that was designed to overcome some of the limitations of the earlier Palmaz-Schatz (PS) stent design. The NIR Vascular Advanced North American (NIRVANA) trial randomized 849 patients with single coronary lesions to treatment with the NIR stent or the PS stent. The study was an “equivalency” trial, designed to demonstrate that the NIR stent was not inferior to (i.e., equivalent or better than) the PS stent, for the primary end point of target vessel failure (defined as death, myocardial infarction, or target vessel revascularization) by 9 months. Successful stent delivery was achieved in 100% versus 98.8%, respectively, with a slightly lower postprocedural diameter stenosis (7% vs. 9%, p ⴝ 0.04) after NIR and PS stent placement, respectively. Major adverse

cardiac events (death, myocardial infarction, repeat target lesion revascularization) were not different at 30 days (4.3% vs. 4.4%). The primary end point of target vessel failure at 9 months was seen in 16.0% of NIR versus 17.2% of PS patients, with the NIR proving to be equal or superior to the PS stent (p <0.001 by test for equivalency). Angiographic restudy in 71% of a prespecified cohort showed no significant difference in restenosis (19.3% vs 22.4%). Thus, the NIR stent showed excellent deliverability with slightly better acute angiographic results and equivalent or better 9-month target vessel failure rate when compared with the PS stent. 䊚2001 by Excerpta Medica, Inc. (Am J Cardiol 2001;87:152–156)

espite its superior performance over balloon angioplasty in a wide variety of applications, the D original 15-mm articulated Palmaz-Schatz (PS; Cor-

limitations has stimulated the development of ⬎20 “second generation” alternative stent designs, which have been evaluated in “stent versus stent” trials that compared each new stent with the previous “gold standard” PS stent in elective native coronary use. Although the PS stent is no longer in mainstream clinical use, and there has been continued evolution in the delivery systems of new stents (and current stent applications extend well beyond those studied in the equivalency trials7), the stent versus stent studies stand as the largest and most controlled body of data on these new stent designs and helps to place them in proper perspective against the well-characterized PS “reference” stent. The NIR stent (Boston Scientific, Natick, Massachusetts) is a second generation stent whose novel multicellular transformable design is engineered to provide maximal flexibility during delivery, yet maximum scaffolding once deployed. Two small registries showed excellent acute procedural success and low rates of angiographic and clinical restenosis.8,9 The NIR Vascular Advanced North American trial, as reported in this study, is the pivotal randomized trial that led to approval of this device by the Federal Drug Adminstration in August 1998.

1– 6

dis, Inc., Johnson & Johnson Interventional Systems, Miami, Florida) coronary stent with its associated delivery sheath has exhibited a number of shortcomings (rigidity, difficult passage through tortuous anatomy, poor scaffolding in the articulation site, the need for multiple stents to treat lesions ⬎15 mm long, and limited radiographic visibility). Recognition of these From the Center for Innovative Minimally Invasive Therapy, and the Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts; University of Rochester Medical School, Rochester, New York; North Ohio Heart Center, Elyria, Ohio; Indiana Heart Institute, Indianapolis, Indiana; Ohio Heart Health Center, Cincinnati, Ohio; SJH Cardiac Cath Associates, Syracuse, New York; Division of Cardiology, North Shore University Hospital, Manhasset, New York; Division of Interventional Cardiology, Lenox Hill Hospital, New York, New York; Cardiovascular Data Analysis Center, Boston, Massachusetts; and Interventional Cardiology, Beth Israel Deaconess Medical Center, Boston, Massachusetts. Manuscript received May 4, 2000; revised manuscript received and accepted July 25, 2000. Address for reprints: Donald S. Baim, MD, Center for Innovative Minimally Invasive Therapy, Brigham and Women’s Hospital, 75 Francis St., Boston, Massachusetts 02115. E-mail: dbaim@partners. org. *A list of the NIRVANA Investigators appears in the Appendix.

152

©2001 by Excerpta Medica, Inc. All rights reserved. The American Journal of Cardiology Vol. 87 January 15, 2001

METHODS

Trial design: The trial was designed to show equivalence in target vessel failure between the 2 stent 0002-9149/01/$–see front matter PII S0002-9149(00)01307-2

arms. Secondary end points included acute success, abrupt closure, 30-day stent thrombosis, angiographic restenosis, target vessel revascularization, and target lesion revascularization. A cohort of 302 consecutive patients (determined by enrollment date) was selected for routine 9-month angiographic follow-up to evaluate the angiographic restenosis end points as well as luminal dynamics (acute gain, late loss, and loss index). Inclusion criteria were: age ⬎18 years, angina or objective evidence of myocardial ischemia, a focal (ⱕ25 mm) de novo lesion (50% to 99% diameter stenosis) in a native coronary artery, suitably sized (visually estimated reference diameter 3.0 to 4.0 mm) for coverage by 1 or 2 PS stents or a single appropriate length NIR stent. Exclusion criteria included involvement of the left main coronary artery or the ostium of a major epicardial coronary artery, the need to stent across a large (⬎2 mm) side branch, a recent myocardial infarction (within 3 days), left ventricular ejection fraction ⬍25%, bleeding, or intolerance to aspirin or ticlopidine. All subjects provided written informed consent under a protocol sanctioned by the local Institutional Review Board at each of the participating hospitals. Study stents: The standard 15-mm PS articulated coronary stent (3.0, 3.5, or 4.0 mm diameter) with the sheath Stent Delivery System (Johnson & Johnson, Warren, New Jersey) was utilized according to the instructions for use contained in the package insert. The NIR stent was an over-the-wire single operator exchange catheter with a balloon expandable stent premounted on the balloon. The NIR stent is constructed from sheets of 316 low carbon stainless steel that are etched into a continuous uniform multicellular design containing either 7 (3.0- and 3.5-mm stents) or 9 (4.0-mm stents) repeating cells around the stent circumference. In their unexpanded configuration, these cells are capable of differential lengthening to confer flexibility. After expansion, these cells stiffen to provide maximal support. For the randomized trial, the stent was premounted on a rapid-exchange balloon (Primo, Boston Scientific) of a length suitable for each stent length (9, 16, and 32 mm). Implantation technique: Stent deployment was performed by slow inflation to 8 to 10 atm. After deflation of the delivery balloon, the delivery catheter was withdrawn, and a high-pressure postdilating balloon sized to match the reference diameter was inserted and inflated to a pressure no less than 12 atm. Intravascular ultrasound (IVUS) was left to the discretion of the operator, and was used in 7.3% of patients (6.5% PS, 8.0% NIR). Concomitant medical therapy: All patients were pretreated with aspirin, and received intravenous heparin sufficient to prolong the activated clotting time to ⬎250 seconds. Arterial sheaths were removed according to each institution’s protocol, usually on the same day as stenting, once the activated clotting time had decreased to ⬍160 seconds. Heparin infusion was generally not reinitiated after sheath removal. Poststent medical therapy involved aspirin 325 mg/day

and ticlopidine 250 mg twice daily for 30 days (the first dose of which was given before or immediately after stent placement).10 Clinical follow-up: Adverse events were tabulated before discharge and clinical follow-up was repeated at 30 days and at 3, 6, and 9 months. All clinical events were adjudicated by a Clinical Events Committee of physicians blinded to stent assignment. Deaths were adjudicated as cardiac unless a clear noncardiac cause was identified. Q-wave myocardial infarction was defined by the presence of new pathologic Q waves in ⱖ 2 contiguous electrocardiographic leads, as determined by the core laboratory. Non–Q-wave myocardial infarction was defined according to the World Health Organization enzymatic criteria (peak CK twice the upper limit of normal, with elevated MB fraction).11 Stent thrombosis included angiographically documented vessel occlusions within 30 days and all cardiac deaths within 30 days that were not clearly due to other causes. Target lesion revascularization required presence of clinical or functional ischemia and ⬎50% diameter stenosis, or a restenotic lesion ⬎70% in the absence of documented ischemia. Follow-up angiography was performed at 6.2 ⫾ 1.2 months (range 2 to 10) in 211 of the 298 eligible patients (71%) in the angiographic subset. Quantitative angiographic analysis: All procedural and follow-up cine angiograms were analyzed by an independent core laboratory blinded to treatment strategy. Standard morphologic criteria were used to characterize baseline lesion complexity and identify the occurrence of angiographic complications.12,13 Lesion length was determined by the “shoulder-to shoulder” axial obstruction extent. Quantitative angiographic analysis was performed using a standard algorithm.14 Binary restenosis was defined as the percent of patients with follow-up diameter stenosis of ⬎50%. Statistical analysis: This study was designed to demonstrate equivalence in the primary end point between the 2 arms. Using the Blackwelder formula to determine the sample size,15 the null hypothesis was constructed so that NIR stent target vessel failure rate was greater than the PS rate plus additional percentage points referred to as the “delta.” The delta for this study was set at 7.5%. All analyses were performed using intent-to-treat samples. Continuous variables were examined using t tests or Wilcoxon nonparametric tests. Binary and polychotomous variables were examined using Fisher’s exact and chi-square tests. Survival estimates were computed using Kaplan-Meier methods and compared using logrank tests. Predictors of binary outcomes (e.g., restenosis) were analyzed using multivariable logistic regression models.16 Continuous measures (e.g., changes in angiographic minimal lumen diameter) were evaluated using multivariable linear regression models.17 Continuous measures are summarized as mean ⫾ SD; frequencies are displayed as counts and percentages. A 2-sided value of p ⬍0.05 was required for statistical significance. All statistical analyses were performed using SAS for Windows (versions 6.10 – 6.12, SAS Institute, Cary, North Carolina).

CORONARY ARTERY DISEASE/COMPARING THE NIR STENT WITH THE PS STENT

153

TABLE 3 Angiographic Results

TABLE 1 Baseline Clinical Characteristics Variable

NIR (n ⫽ 418)

PS (n ⫽ 430)

Age (mean) (yrs) Women Diabetes mellitus Cigarette smoking Dyslipidemia requiring treatment Prior coronary bypass Prior myocardial infarction Prior restenosis Unstable angina Left ventricular ejection fraction

62 ⫾ 11 30% 23% 31% 67% 7% 42% 12% 75% 55 ⫾ 11%

62 ⫾ 11 32% 22% 29% 61% 8% 37% 10% 73% 55 ⫾ 11%

TABLE 2 Baseline Angiographic Findings

Coronary artery stented Left anterior descending Left circumflex Right Baseline characteristics Eccentric Calcification Bend ⬎45° Thrombus Total occlusions Lesion morphology (ACC/AHA) A B1 B2 C

NIR (n ⫽ 418)

PS (n ⫽ 430)

42% 21% 37%

40% 22% 38%

41% 20% 11% 3% 1%

43% 18% 9% 3% 1%

9% 23% 52% 17%

6% 28% 50% 15%

Baseline Reference diameter (mm) Angiographic subset Lesion minimal lumen diameter (mm) Angiographic subset Diameter stenosis (%) Angiographic subset Lesion length (mm) Angiographic subset Post procedure In-stent minimal lumen diameter (mm) Angiographic subset Acute gain Angiographic subset In-stent diameter stenosis (%) Angiographic subset Follow-up† Follow-up minimal lumen diameter (mm) Late loss Loss index‡ Restenosis (%, in-stent)

NIR (n ⫽ 418)

PS (n ⫽ 430)

2.97 ⫾ 0.52 2.98 ⫾ 0.51 1.04 ⫾ 0.43

3.03 ⫾ 0.52 2.96 ⫾ 0.52 1.08 ⫾ 0.42

1.04 65 65 13.3 13.4

⫾ ⫾ ⫾ ⫾ ⫾

0.44 13 14 6.9 7.1

2.78 ⫾ 0.44 2.74 1.75 1.73 7 8

⫾ ⫾ ⫾ ⫾ ⫾

0.41 0.51 0.44 11* 11

1.09 64 63 13.3 13.4

⫾ ⫾ ⫾ ⫾ ⫾

0.42 13 14 7.1 7.6

2.79 ⫾ 0.43 2.73 1.71 1.64 9 8

⫾ ⫾ ⫾ ⫾ ⫾

0.40 0.51 0.49 12 11

2.00 ⫾ 0.75

1.90 ⫾ 0.76

0.80 ⫾ 0.61 0.48 ⫾ 0.43 19.3

0.85 ⫾ 0.60 0.56 ⫾ 0.44 22.4

*p ⬍0.05 for NIR versus PS; †angiographic follow-up subset only; ‡loss index ⫽ slope of the regression of late loss versus acute gain ⫾ SE. Patients who failed to have a stent delivered were not included in the postprocedure measurements.

ACC ⫽ American College of Cardiology; AHA ⫽ American Heart Association.

RESULTS

Baseline characteristics: A total of 849 patients were randomized between January and April 1997 to receive either the PS or the NIR stent, at 41 participating centers in the United States (see Appendix). One patient was excluded due to withdrawal of consent. Baseline clinical and angiographic characteristics were similar between the 2 stent cohorts (Tables 1 and 2). Procedural results: Final balloon:artery ratio (1.13 ⫾ 0.16 vs 1.14 ⫾ 0.16) and mean final postdilation pressure (15.5 ⫾ 3.3 vs 16.6 ⫾ 3.1 atm) were similar in the NIR and PS arms, respectively. Device success (achievement of a residual diameter stenosis ⬍50%, using the assigned device only) was 99.5% versus 97.9% for the NIR and PS stents, respectively. Failure of at least 1 attempted device placement occurred in 4 NIR patients (1.0%) versus 9 PS patients (2.1%). At least 1 NIR stent was delivered in every patient randomized to NIR. Among PS patients, however, no assigned stent could be delivered in 5 patients (1.2%), so that these patients underwent balloon angioplasty alone or received another approved stent type. Procedural success (device success, without in-hospital death, myocardial infarction, or emergency bypass surgery) was also high (95.4% for the NIR stent and 94.3% for the PS stent). Angio154 THE AMERICAN JOURNAL OF CARDIOLOGY姞

VOL. 87

FIGURE 1. Baseline, acute, and follow-up angiographic results. There was no significant difference in percent diameter stenosis at baseline (Pre). After stent placement (Post), there was a small, but significant improvement in postprocedural percent diameter stenosis for NIR. At 9-month angiographic follow-up (F/U, angiographic subset only) between NIR and PS patients, there was a nonsignificantly lower diameter stenosis for NIR.

graphic results (Table 3, Figure 1) show a small but statistically significant reduction in postprocedural instent residual stenosis (7 ⫾ 11% vs 9 ⫾ 12%, p ⫽ 0.04) with the NIR stent. Clinical follow-up (Table 4): At 30 days, composite major adverse events were similar (4.3% vs 4.4%, p ⬎0.20) for the NIR and PS stents, respectively. At 9 months, with follow-up available in 96% of eligible patients (97% NIR, 96% PS), the incidence of major JANUARY 15, 2001

imal lumen diameter (odds ratio 0.46/mm, p ⫽ 0.0006). Angiographic follow-up: Routine 9-month angiographic follow-up (Table 3, Figure 1) was completed in 71% of the 298 patients eligible for follow-up (78% of NIR and 66% of PS patients). Angiographic restenosis was nonsignificantly lower for NIR (19.3% vs 22.4%). Multivariable logistic regression modeling showed that the independent determinants of the probability of restenosis were: total stent length (odds ratio 1.09/mm, p ⫽ 0.0001), history of diabetes (odds ratio 4.58, p ⫽ 0.0005), poststent minimal lumen diameter (odds ratio 0.19/mm, FIGURE 2. Event-free survival. Kaplan-Meier estimate of freedom from target vessel p ⫽ 0.003), left anterior descending failure (TVF). The observed difference in the Kaplan-Meier estimate after 9 months in target vessel (odds ratio 2.95, p ⫽ favor of NIR (16.2% vs 17.6%) was not statistically significant (see text for discussion 0.012), and age (odds ratio 1.05/ of equivalency). year, p ⫽ 0.021). After adjustment for these variables, there was no independent effect of stent type on the late angiographic outcomes. TABLE 4 Clinical Follow-up As in prior studies, the performance of routine NIR PS angiographic follow-up on a subset of NIRVANA (n ⫽ 418) (n ⫽ 430) patients influenced clinical restenosis parameters.18,19 Unadjudicated target lesion revascularization by 1 30-d cumulative events Any MACE 4.3% 4.4% year was 15.3% for patients with routine angiographic Death 0.0% 0.2% follow-up versus 10.6% for patients with clinical folMyocardial infarction 4.1% 3.7% low-up only (p ⫽ 0.049). Even with adjudication by Q wave 0.5% 0.9% the clinical events committee (according to rules to Non-Q-wave 3.6% 2.8% Emergency coronary bypass surgery 0.2% 0.0% discriminate between clinically indicated vs clinically Target lesion revascularization 1.4% 1.4% nonindicated revascularization), the effect of undergoCoronary artery bypass surgery 0.2% 0.2% ing routine angiographic follow-up on target lesion Coronary angioplasty (only) 1.2% 1.2% revascularization was only partially corrected (13.7% Stent thrombosis 0.5% 0.5% vs 10.0%, p ⫽ 0.11). Vascular 0.5% 0.9% Other major bleeding 9-mo cumulative events Death Myocardial infarction Q wave Non–Q-wave Target lesion revascularization Coronary angioplasty (only) Coronary artery bypass surgery Target vessel revascularization Target vessel failure

0.5%

0.7%

1.0% 4.8% 0.7% 4.1% 9.6% 7.2% 2.4% 12.2% 16.0%

0.9% 4.2% 0.9% 3.3% 11.6% 8.6% 3.0% 13.4% 17.2%

p ⬎0.20 for all. MACE ⫽ Major adverse cardiac event (death, myocardial infarction, or target lesion revascularization).

events remained low in both groups (Figure 2). The prespecified test for equivalency yielded a strong significant result (p ⫽ 0.0003), rejecting the null hypothesis (i.e., the NIR is inferior to the PS stent), and accepting the alternate hypothesis (i.e., the NIR stent is equal or superior to the PS stent). Multivariable logistic regression modeling showed that the only independent determinants of the probability of target vessel failure were: total stent length (odds ratio 1.03/mm, p ⫽ 0.0001) and poststent min-

DISCUSSION

Equivalency trials versus usual difference trials:

Analysis of the outcome of stent versus stent trials, such as NIRVANA, requires a statistical approach other than that used in the usual “difference” trial, where the goal is to show that a new therapy is superior to the prior standard therapy.20 Although failure to show a statistically significant (p ⬍0.05) difference between the experimental treatment and a control group in a difference trial is generally reported as indicating “no significant difference” between the 2 groups, this is not the same as demonstrating that the 2 treatments are statistically equivalent (for example, the difference trial may simply have been underpowered to find the expected differences, even if they existed). In contrast, an “equivalence trial” such as NIRVANA takes as its null hypothesis that the new treatment is worse than the standard therapy, in regard to the primary end point. Finding a significant p value (⬍0.05) in the equivalence design forces rejection of that null hypothesis (inferiority), and acceptance of the alternative hypothesis—that the new stent is equivalent or superior to the standard stent.

CORONARY ARTERY DISEASE/COMPARING THE NIR STENT WITH THE PS STENT

155

Acute procedural results and clinical outcomes: NIRVANA shows that the NIR stent (9, 16, or 32 mm) was more deliverable than the 15-mm articulated PS stent with its sheathed delivery system. All lesions randomized to the NIR stent were ultimately treated with ⱖ1 study stent. In contrast, in 1.2% of PS patients, a study stent could not be delivered. There were comparable composite major adverse cardiac event rates between the NIR and PS stents at 30 days. Of these events, most represented non–Qwave myocardial infarctions, with few deaths, Q-wave myocardial infarctions, or subacute stent thromboses. Angiographic follow-up: Angiographic follow-up data in 71% of eligible patients showed a nonsignificant reduction in binary restenosis for NIR. Any benefit of the NIR stent, however, flowed from its better acute result (rather than any reduction in proliferative stimulation), because the loss indexes for the 2 stents were not significantly different. It should also be noted that incomplete ascertainment of follow-up angiographic data may have resulted in an overestimation of angiographic restenosis, and that differences in follow-up may have biased these results in favor of the NIR stent.21 Clinical follow-up: At 9 months there were nonsignificant reductions in target lesion revascularization, target vessel revascularization, and target vessel failure for NIR. When subjected to the predefined statistical test for equivalency, there was a strongly significant value (p ⬍0.001) for rejecting the possibility that the NIR was worse than the PS stent (by ⬎7.5%), and concluding that the NIR was equivalent to or better than the earlier standard. Effect of routine angiographic follow-up on clinical restenosis: NIRVANA supports the finding from pre-

vious trials,18,19 that patients who are subjected to routine angiographic follow-up are more likely to undergo repeat target lesion revascularization. This difference persisted even after adjudication by a Clinical Events Committee blinded to stent assignment and again showed the potential pitfall of comparing clinical end point rates directly between trials that did and did not use routine angiographic follow-up in their whole cohort or a substantial subset.

APPENDIX In addition to the study investigators, the following investigators and research coordinators (listed in descending order of enrollment) participated in the NIR Vascular Advanced North American trial: Mount Sinai Hospital, New York, NY: S. Sharma, D. Ratner; Emory University Hospital, Atlanta, GA: J. Douglas, R. Askins; Good Samaritan Hospital, Phoenix, AZ: N. Laufer, L. Hill; Sharp Memorial Hospital, San Diego, CA: M. Buchbinder, V. Nassar; Lenox Hill Hospital, New York, NY: J. Moses, N. Cohen; The Methodist Hospital, Houston, TX: A. Raizner, J. Joseph; Washington Hospital Center, Washington, DC: M. Leon, L. Gierlach; New York Hospital & Medical Center of Queens, Flushing, NY: S. Wong, M.C. Boileau; Cleveland Clinic, Cleveland, OH: S. Ellis, N. Juran; Texas Heart Institute, Houston, TX: D. Fish, M. Harlan; Mercy Hospital Medical Center, Des Moines, IA: L. Iannone, J. Greene; Oschner Clinic, New Orleans, LA: S. Ramee, N. McCarthy; Miriam Hospital, Providence, RI: P. Gordon, N. Wright; Johns Hopkins Hospital, Baltimore: MD, J. Brinker, V. Coombs; Yale New Haven Hospital, New Haven, CT: M. Cleman, J. Davey; Riverside Methodist Hospital, Columbus, OH: B. George, K. Rusk; Beth Israel Deaconess Medical Center, Boston, MA: D. Baim, P. Flatley; Jewish Heart & Lung Institute, Louisville, KY: D. Senior, V. Miracle; Mercy General Hospital, Sacramento, CA: R. Low, K. Zirbel; Rhode Island Hospital, Providence, RI: D. Williams, J. Muratori; Scripps Clinic, La Jolla, CA: P. Teirstein, K. Jaconette; The Sanger Clinic, Charlotte, NC: R. Bersin, J. Lamb; Arizona Heart Institute, Phoenix, AZ: R. Strumpf or R. Heuser, S. Spooner; St. Joseph’s Hospital of Atlanta,

156 THE AMERICAN JOURNAL OF CARDIOLOGY姞

VOL. 87

Atlanta, GA: C. Cates, T. Riach; Mid-American Heart Institute, Kansas City, MO: B. McAllister, P. Rosson; Swedish Medical Hospital, Seattle, WA: M. Reisman, V. Harms; Barnes-Jewish Hospital, St. Louis, MO: J. Lasala, A. Campbell; William Beaumont Hospital, Royal Oak, MI: R. Safian, D. Davey; Temple University, Philadelphia, PA: J. Burke or E. Deutsch, B. Flynn; Hartford Hospital, Hartford, CT: R. McKay, D. Murphy; Massachusetts General Hospital, Boston, MA: I. Palacios, C. Cunneen; Georgetown University Medical Center, Washington, DC: D. Diver, J. Gannuscio; Allegheny General Hospital, Pittsburgh, PA: D. Lasorda, E. Romano; Valley Hospital Medical Center, Las Vegas, NV: C. Fonte, S. Zach; New York Beth Israel Medical Center, New York, NY: J. Wilentz, D. McDermott; University of North Carolina Hospitals, Chapel Hill, NC: G. Dehmer, M. Fisher. Data Coordinating Center: R.E. Kuntz, K.K.L. Ho, D.E. Cutlip, C. Senerchia, T. DeFeoFraulini, M. Fitzpatrick, A. Lanoue, C. Rizzitano. Angiographic Core Laboratory: J.J.Popma, A. Lansky, R. Mehran, J. Saucedo. Clinical Events Committee: J.P. Carrozza, D.J. Cohen, L.M. Epstein, J.P. Kannam, W.J. Manning, J. Markis, J. P. Oettgen, M. Simons. ECG Core Laboratory: A.L. Goldberger. 1. Fischman DL, Leon MB, Baim DS, Schatz RA, Savage MP, Penn I, Detre K, Veltri L, Ricci D, Nobuyoshi M, et al. A randomized comparison of coronarystent placement and balloon angioplasty in the treatment of coronary artery disease. Stent Restenosis Study Investigators. N Engl J Med 1994;331:496 –501. 2. Serruys PW, de Jaegere P, Kiemeneij F, Macaya C, Rutsch W, Heyndrickx G, Emanuelsson H, Marco J, Legrand V, Materne P, et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. Benestent Study Group. N Engl J Med 1994;331:489 – 495. 3. Savage MP, Douglas JSJ, Fischman DL, Pepine CJ, King SBI, Werner JA, Bailey SR, Overlie PA, Fenton SH, Brinker JA, Leon MB, Goldberg S. Stent placement compared with balloon angioplasty for obstructed coronary bypass grafts. Saphenous Vein De Novo Trial Investigators. N Engl J Med 1997;337: 740 –747. 4. Simes P, Golf S, Myreng Y, Molstad P, Emanuelsson H, Albertsson P, Brekke M, Mangschau A, Endresen K, Kjekshus J. Stenting in chronic coronary occlusion (SICCO): a randomized trial of adding stent implantation after successful angioplasty. J Am Coll Cardiol 1996;28:1444 –1451. 5. Suryapranata H, van’t Hof AW, Hoorntje JC, de Boer MJ, Zijlstra F. Randomized comparison of coronary stenting with balloon angioplasty in selected patients with acute myocardial infarction. Circulation 1998;97:2502–2505. 6. Versaci F, Gaspardone A, Tomai F, Crea F, Chiariello L, Gioffre P. A comparison of coronary-artery stenting with angioplasty for isolated stenosis of the proximal left anterior descending coronary artery. N Engl J Med 1998;336: 817– 822. 7. Linnemeier T. The Rx Multi-Link stent parallel registries to the ASCENT trial. J Inv Cardiol 1998;10(suppl. B):55B–56B. 8. Lau K, He Q, Ding Z, Quek S, Johan A. Early experience with the NIR intracoronary stent. Am J Cardiol 1998;81:927–929. 9. Zheng H, Corcos Y, Favreau X, Pentousis D, Guerin Y, Ouzan J, Toussaint M. Preliminary experience with the NIR coronary stent. Cathet Cardiovasc Diagn 1998;43:153–158. 10. Leon MB, Baim DS, Popma JJ, Gordon PC, Cutlip DE, Ho KKL, Giambartolomei A, Diver DJ, Lasorda DM, Williams DO, Pocock SJ, Kuntz RE. A clinical trial comparing three anti-thrombotic drug regimens after coronary artery stenting. N Engl J Med 1998;338:1665–1671. 11. WHO Monica Project. Monica Manual. Revised Edition. Geneva, Switzerland: Cardiovascular Diseases Unit, WHO, 1990. 12. Ellis SG, Vandormael MG, Cowley MJ, Di Sciascio G, Deligonul U, Topol EJ, Bulle TM. Coronary morphologic and clinical determinants of procedural outcome with angioplasty for multivessel coronary disease. Implications for patient selection. Multivessel Angioplasty Prognosis Study Group. Circulation 1990;82:1193–1202. 13. Popma J, Bashore T. Qualitative and quantitative coronary angiography. In: Topol E, ed. Interventional Cardiology. Philadelphia: WB Saunders, 1993:1052– 1068. 14. van der Zwet P, Reiber J. A new approach for the quantification of complex lesion morphology: the gradient field transform; basic principles and validation results. J Am Coll Cardiol 1994;24:216 –224. 15. Blackwelder W. “Proving the null hypothesis” in clinical trials. Contr Clin Trials 1982;3:343–352. 16. SAS Institute. The Logistic Procedure. SAS/STAT User’s Guide. Version 6. Cary, NC: SAS Institute Inc., 1990:1071–1134. 17. SAS. The REG Procedure. SAS/STAT User’s Guide. Version 6. Cary, NC: SAS Institute Inc., 1990:1351–1456. 18. Serruys PW, van Hout B, Bonnier H, Legrand V, Garcia E, Macaya C, Sousa E, van der Giessen W, Colombo A, Seabra-Gomes R, et al. Randomized comparison of implantation of heparin-coated stents with balloon angioplasty in selected patients with coronary artery disease (Benestent II). Lancet 1998;352: 673– 681. 19. Cutlip D, Ho K, Kuntz R, Baim D. Influence of routine angiographic follow-up on clinical restenosis outcome in the ASCENT trial (abstr). J Am Coll Cardiol 1998;31(suppl.):139A. 20. Ware JH, Antman EM. Equivalence trials (editorial). N Engl J Med 1997; 337:1159 –1161. 21. Kuntz RE, Keaney KM, Senerchia C, Baim DS. A predictive method for estimating the late angiographic results of coronary intervention despite incomplete ascertainment. Circulation 1993;87:815– 830.

JANUARY 15, 2001