- Email: [email protected]
Impact of Routine Angiographic Follow-Up After Percutaneous Coronary Intervention With Drug-Eluting Stents in the SPIRIT III Randomized Trial at Three Years
Impact of Routine Angiographic Follow-Up After Percutaneous Coronary Intervention With Drug-Eluting Stents in the SPIRIT III Randomized Trial at Three Years Alexandra J. Lansky, MDa,*, Somjot S. Brar, MDb, Manejeh Yaqub, MDc, Poornima Sood, MD, MBAc, Robert J. Applegate, MDd, Dana Lazar, MDa, Ivana Jankovic, MDa, James B. Hermiller, MDe, Kai Koo, PhDc, Krishnankutty Sudhir, MD, PhDc, and Gregg W. Stone, MDf Routine angiographic follow-up after bare-metal stent implantation has been associated with an increase in coronary revascularization. The impact of angiographic follow-up after drug-eluting stent placement remains poorly characterized. The prospective, randomized, single-blinded SPIRIT III trial assigned patients to the everolimus-eluting stent or the paclitaxel-eluting stent (PES). Major adverse cardiovascular events (cardiac death, myocardial infarction, and ischemia-driven target lesion revascularization [ID-TLR]) at 3 years were assessed by angiographic versus clinical-only follow-up at 8 months ⴞ 28 days and a landmark survival analysis from 9 months to 3 years. Of 1,002 patients, 564 patients were assigned to angiographic follow-up at 8 months ⴞ 28 days and 438 patients underwent clinical follow-up alone. Three-year major adverse cardiovascular event rates were 10.6% in the angiographic group and 12.0% in the clinical follow-up group (p ⴝ 0.64). Ischemiadriven revascularization increased twofold at 9 months, but no difference was noted in ID-TLR for either device. Non–ID-TLR was significantly higher in patients in the angiographic group (4.5% vs 1.0%, p ⴝ 0.002), a difference resulting from PES (9.1% vs 0.7%, p ⴝ 0.0007) rather than everolimus-eluting stent (2.2% vs 1.1%, p ⴝ 0.36) treatment. The landmark analysis showed no significant differences between the angiographic and clinical follow-up groups from 9 months to 3 years of major clinical outcomes. In conclusion, routine angiographic follow-up in SPIRIT III did not increase rates of ID-TLR compared to clinical follow-up alone. Despite higher nonischemia-driven revascularization rates with angiographic follow-up of patients with PESs, none of the safety end points were adversely affected. © 2012 Elsevier Inc. All rights reserved. (Am J Cardiol 2012;110:21–29) The benefits of percutaneous coronary intervention compared to balloon angioplasty have been tempered by restenosis.1–5 Although bare-metal stents (BMSs) decreased the risk of restenosis, the introduction of drug-eluting stents (DESs) resulted in a marked 60% to 80% decrease in restenosis compared to balloon angioplasty and BMSs.6 –12 Despite the widespread adoption of DESs,13,14 controversy
a
Yale University School of Medicine, New Haven, Connecticut; bKaiser Permanente Medical Center, Los Angeles, California; cAbbott Vascular, Santa Clara, California; dWake Forest University Health Sciences, Winston-Salem, North Carolina; eSt. Vincent’s Heart Center of Indiana, Indianapolis, Indiana; fColumbia University Medical Center, Cardiovascular Research Foundation, New York, New York. Manuscript received July 7, 2011; revised manuscript received and accepted February 21, 2012. This work was funded by Abbott Vascular, Santa Clara, California. Dr. Applegate reports serving on scientific advisory boards and honoraria/ consulting for Abbott Vascular and receiving research funding from Abbott Vascular. Dr. Hermiller reports serving as a consultant for Abbott Vascular and Boston Scientific Corp., Natick, Massachusetts. Dr. Stone reports serving on scientific advisory boards for and honoraria from Abbott Vascular and Boston Scientific Corp. and as a consultant to Medtronic, Santa Rosa, California. *Corresponding author: Tel: 203-737-2142; fax: 203-737-6118. E-mail address: [email protected] (A.J. Lansky). 0002-9149/12/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2012.02.040
remains on the clinical impact of protocol-mandated angiographic follow-up in DES clinical trials, which may preferentially accentuate rates of repeat revascularization and negatively affect safety outcomes.15 Follow-up angiography has been associated with exaggerated rates of revascularization, ostensibly resulting from the “oculostenotic reflex,” a term describing revascularization using percutaneous coronary intervention owing to anatomic lesion severity independent of clinical or physiologic evidence of ischemia. Although this phenomenon is well recognized using balloon angioplasty and BMSs, few investigations have studied the impact of routine angiographic follow-up with DESs. This study aimed to assess the impact of protocol-mandated angiographic follow-up in patients enrolled in A clinical Evaluation of the Investigational Device XIENCE V Everolimus Eluting Coronary Stent System (EECSS) in the Treatment of Subjects With de Novo Native Coronary Artery Lesions (SPIRIT III) trial. Methods The SPIRIT III trial was a prospective, multicenter, randomized, controlled, single-blinded trial in which 1,002 patients were randomized in a 2:1 ratio to the everolimuseluting stent (EES; XIENCE V, Abbott Vascular, Santa www.ajconline.org
22
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics, three-year thienopyridine compliance, and procedural and angiographic characteristics of study population Angiographic Group (n ⫽ 564) Baseline characteristics Demographics Age (years), mean ⫾ SD Men Hypertension Hypercholesterolemia Diabetes mellitus Any Requiring insulin Smoker, current Target vessel Left anterior descending coronary artery Left circumflex coronary artery Right coronary artery Left main coronary artery (protected) Number of narrowed coronary arteries 1 2 ⱖ3 American College of Cardiology/America Heart Association lesion class A B1 B2 C Thienopyridine use At 180 days At 270 days At 365 days At 730 days At 1,095 days Procedural and angiographic characteristics Procedural variables, mean ⫾ SD Number of lesions Number of stents per patient Number of stents per lesion Maximum stent diameter per lesion (mm) Total stent length per lesion (mm) Maximum pressure (atm) Glycoprotein IIb/IIIa inhibitors used
Clara, California) or the paclitaxel-eluting stent (PES; TAXUS Express2, Boston Scientific, Natick, Massachusetts).12 In brief, patients with 1 de novo lesion or 2 de novo coronary lesions with reference vessel diameters of 2.5 to 3.75 mm and lesions ⱕ28 mm in length were eligible for randomization. Telephone randomization was performed in blocks of 3 and 6 patients stratified by diabetes status, 2-vessel treatment, and study site. EESs were available in 2.5-, 3.0-, and 3.5-mm diameters with lengths of 8, 18, and 28 mm. PESs were available in 2.5- to 3.5-mm diameters and lengths ranging from 8 to 32 mm. Operators were instructed to choose a stent length to cover approximately 3 mm of nondiseased tissue at the proximal and distal stent edges. When multiple stents were necessary, 1 to 4 mm of stent overlap was recommended. Operators were not blinded to randomization; however, patients and staff involved in follow-up assessments remained blinded throughout follow-up. The protocol-mandated angiographic fol-
Nonangiographic Group (n ⫽ 438)
p Value
63.3 ⫾ 10.4 387/563 (68.7%) 419/562 (74.6%) 400/554 (72.2%)
62.8 ⫾ 10.4 300/438 (68.5%) 336/438 (76.7%) 323/431 (74.9%)
0.41 0.95 0.46 0.35
160/562 (28.5%) 38/56 (26.8%) 121/551 (22.0%)
29.7% (130/437) 7.3% (32/437) 106/433 (24.5%)
0.67 0.80 0.36
269/647 (41.6%) 182/647 (28.1%) 196/647 (30.3%) 0/647 (0.0%)
212/503 (42.1%) 138/503 (27.4%) 151/503 (30.0%) 2/503 (0.4%)
0.86 0.84 0.95 0.19
378/563 (67.1%) 136/563 (24.2%) 49/563 (8.7%)
276/438 (63.0%) 117/438 (26.7%) 44/438 (10.0%)
0.18 0.38 0.51
44/640 (6.9%) 223/640 (34.8%) 243/640 (38.0%) 130/640 (20.3%)
35/50 (17.0%) 188/501 (37.5%) 199/501 (39.7%) 79/501 (15.8%)
1.00 0.35 0.58 0.05
524/559 (93.7%) 408/549 (74.3%) 372/543 (68.5%) 277/515 (53.8%) 253/512 (49.4%)
411/432 (95.1%) 337/426 (79.1%) 315/417 (75.5%) 263/410 (64.1%) 225/402 (56.0%)
0.41 0.09 0.02 0.002 0.05
647 1.3 ⫾ 0.6 1.2 ⫾ 0.4 3.01 ⫾ 0.37 22.4 ⫾ 8.0 14.8 ⫾ 2.8 150/563 (26.6%)
508 1.3 ⫾ 0.6 1.1 ⫾ 0.4 3.03 ⫾ 0.39 22.3 ⫾ 8.6 15.0 ⫾ 2.8 116/438 (26.5%)
0.38 0.09 0.31 0.89 0.28 1.00
low-up was specified at 8 months for the first 564 randomized patients in the study. A landmark analysis was performed from the last day of the 8th-month angiographic follow-up window (268 days) or from 9 months to 3 years. Clinical end points used in the SPIRIT III trial have been previously defined.12 In brief, major adverse cardiovascular events were defined as the composite of cardiac death, myocardial infarction, and ischemia-driven target lesion revascularization (ID-TLR). Myocardial infarction was defined as the development of new pathologic Q waves ⱖ0.4 second in duration in ⱖ2 contiguous leads or as an increase in creatinine phosphokinase-MB. Revascularization was categorized as ID or non- ID. ID revascularization required (1) a positive functional study, or (2) the presence of symptoms and a ⬎50% diameter stenosis by independent core laboratory, or (3) in the absence of symptoms or a positive functional study required ⬎70% diameter stenosis by an independent core
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
23
Table 2 Outcomes of entire cohort at three years Result/Outcome Cardiac death Myocardial infarction Q wave Non-Q wave Stent thrombosis (Academic Research Consortium) Definite Probable Definite or probable Target lesion revascularization, all Ischemia driven Nonischemia driven Target vessel revascularization (nontarget lesion revascularization), all Ischemia driven Nonischemia driven Bleeding complications
Angiographic Group (n ⫽ 564)
Nonangiographic Group (n ⫽ 438)
p Value
9/522 (1.7%) 21/522 (4.0%) 2/522 (0.4%) 19/522 (3.6%)
7/412 (1.7%) 23/412 (5.6%) 3/412 (0.7%) 20/412 (4.9%)
1.00 0.28 0.66 0.41
4/523 (0.8%) 2/523 (0.4%) 6/523 (1.1%) 55/532 (10.3%) 37/532 (7.0%) 24/532 (4.5%) 49/532 (9.2%) 43/532 (8.1%) 7/532 (1.3%) 23/515 (4.5%)
4/408 (1.0%) 3/408 (0.7%) 7/408 (1.7%) 31/416 (7.5%) 27/416 (6.5%) 4/416 (1.0%) 27/416 (6.5%) 26/416 (6.3%) 3/416 (0.7%) 25/405 (6.2%)
0.74 0.66 0.58 0.14 0.80 0.002 0.15 0.31 0.53 0.30
Figure 1. Patient flow in the SPIRIT III randomized clinical trial through 3 years depicting split of patients treated with everolimus-eluting or paclitaxeleluting stents into an angiographic versus a nonangiographic group.
laboratory. ID revascularization was further adjudicated by the angiographic core laboratory based on the location of recurrence as ID-TLR and ID target vessel revascularization (non-TLR or revascularization remote from the original target site). Non-ID-TLR was defined as revascularization in the absence of any of the previously described criteria. Protocol-defined stent thrombosis was (1) an acute coronary syndrome with angiographic evidence of thrombus within or adjacent to a previously treated lesion, or (2), in the absence of angiographic evidence, any unexplained death or ST-segment elevation myocardial infarction, or (3) new Q waves in the distribution of the target lesion within 30 days of the index percutaneous coronary intervention. Stent thrombosis was also
independently adjudicated according to the Academic Research Consortium definition. All clinical events were adjudicated by an independent clinical events committee (Harvard Clinical Research Institute).16 All baseline, follow-up, and event-related angiograms were analyzed by an independent angiographic core laboratory (Cardiovascular Research Foundation, New York, New York) by analysts blinded to treatment allocation using standard definitions and methods for all quantitative measurements using Medis software (Medis, Leiden, the Netherlands). All outcomes are reported at 3 years from the index percutaneous coronary intervention and the 9-month (268 days, upper window of the 8th month) landmark date and are based on the intent-to-treat principle.
24
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Cumulative incidence of major adverse cardiovascular events (MACEs) in patients undergoing routine angiographic follow-up and those not scheduled for angiographic follow-up. (A) Time from index percutaneous coronary intervention to 3 years. (B) Nine-month/268-day (last day of 8th month window) landmark to 3 years. Hazard ratios (HRs) and log-rank p values are calculated at 268 and 1,123 days, respectively.
The cohort was divided into patients undergoing protocol-mandated angiography at 8 months and those undergoing standard clinical follow-up. During the index percutaneous coronary intervention, each patient was treated with an EES or a PES. Landmark survival analysis was used to determine whether there were differences in the primary outcome between routine angiographic and standard clinical follow-up groups after 268 days (last day of 8th month angiographic follow-up window) or the 9-month landmark. Patients in the standard clinical follow-up group who died, developed myocardial infarction, or underwent repeat revascularization within 268 days
of the index percutaneous coronary intervention were excluded from the landmark analysis. Similarly, patients in the routine angiographic group who died, developed myocardial infarction, or underwent repeat revascularization before the scheduled angiographic follow-up were excluded from the landmark analysis. Continuous variables are reported as mean ⫾ SD and were compared by t test. Categorical variables are reported as absolute values and percentages with comparisons made with the Fisher exact test. The Kaplan–Meier method was used to construct survival curves for the primary end point of a major adverse cardiovascular event (cardiac death,
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
25
Figure 3. Cumulative incidence of cardiac death and myocardial infarction (MI) in patients undergoing routine angiographic follow-up and those not scheduled for angiographic follow-up. (A) Time from index percutaneous coronary intervention to 3 years. (B) Nine-month/268-day (last day of 8th month window) landmark to 3 years. Hazard ratios and log-rank p values are calculated at 268 and 1,123 days, respectively. Other abbreviation as in Figure 2.
myocardial infarction, or ID-TLR) and its components. Comparisons between groups were made with log-rank test. Cox proportional hazards models were used to generate hazard ratios and 95% confidence intervals for the study outcomes. The proportionality assumption of the proportional hazards model was assessed and satisfied using log(⫺log) plots. All tests are 2-tailed, with differences reported as statistically significant if the p value was ⬍0.05. All analyses were performed with SAS 9.1 (SAS Institute, Cary, North Carolina).
Results The SPIRIT III trial enrolled 1,002 patients at 65 United States sites from June 2005 through March 2006. Patients were randomized to receive EESs (n ⫽ 669) or PESs (n ⫽ 333). In total 564 randomized patients were assigned to undergo coronary angiography at 8 months. Of these, 435 (77%) underwent angiography as planned at 240 ⫾ 28 days or had a previous angiogram that qualified for analysis. Angiographic and clinical follow-up groups were well
26
The American Journal of Cardiology (www.ajconline.org)
Figure 4. Cumulative incidence of ischemia-driven revascularization, composite of ischemia-driven target lesion revascularization, and ischemia-driven target vessel revascularization. (A) Time from index percutaneous coronary intervention to 3 years. (B) Nine-month landmark to 3 years. Hazard ratios and log-rank p values are calculated at 268 and 1,123 days respectively. Other abbreviation as in Figure 2.
matched for cardiac risk factors, location, and severity of coronary artery disease (Table 1) and for procedural and angiographic characteristics (Table 1). All patients received a thienopyridine at discharge (Table 1), with a shorter mean duration of therapy in the angiographic compared to the clinical follow-up group (682 ⫾ 398 vs 750 ⫾ 376 days, p ⫽ 0.006, respectively). Thienopyridine use was similar at 6 months (180 days) between groups (Table 1). At 9 months, 1 year, 2 years, and 3 years, clopidogrel use was greater in the clinically followed group. In contrast, aspirin use was similar between the angiographic and clinical follow-up groups at all time points, with a mean duration of aspirin
therapy of 999 ⫾ 241 days in the angiographic group and 997 ⫾ 253 days in the clinical follow-up group (p ⫽ 0.92). Follow-up at 3 years was available in 934 patients (93.2% of study cohort), with similar follow-up rates between groups (Table 2, Figure 1). There was no significant difference in 3-year major adverse cardiovascular events between the angiographic and clinical follow-up groups (Figure 2), and the composite of cardiac death and myocardial infarction was also similar (Figure 3). There were no differences between the 2 groups in myocardial infarction, Academic Research Consortium definite or probable stent thrombosis, or bleeding complications (Table 2).
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
27
Table 3 Nine-month (268-day) landmark analysis outcomes at three years Result/Outcome Cardiac death Myocardial infarction Q wave Non-Q wave Stent thrombosis (Academic Research Consortium) Definite Probable Definite or probable Target lesion revascularization, all Ischemia driven Nonischemia driven Target vessel revascularization (nontarget lesion revascularization), all Ischemia driven Nonischemia driven Bleeding complications
Angiographic Group (n ⫽ 564)
Nonangiographic Group (n ⫽ 438)
p Value
5/462 (1.1%) 9/462 (1.9%) 1/462 (0.2%) 8/462 (1.7%)
6/386 (1.6%) 12/386 (3.1%) 3/386 (0.8%) 9/386 (2.3%)
0.56 0.38 0.34 0.63
1/516 (0.2%) 1/516 (0.2%) 2/516 (0.4%) 17/470 (3.6%) 13/470 (2.8%) 5/470 (1.1%) 19/470 (4.0%) 17/470 (3.6%) 2/470 (0.4%) 9/459 (2.0%)
3/406 (0.7%) 3/406 (0.7%) 6/406 (1.5%) 19/390 (4.9%) 17/390 (4.4%) 2/390 (0.5%) 16/390 (4.1%) 16/390 (4.1%) 2/390 (0.5%) 5/381 (1.3%)
0.33 0.33 0.15 0.40 0.26 0.47 1.00 0.72 1.00 0.59
Figure 5. Ischemia-driven revascularization rates for patients treated with everolimus-eluting stents. CI ⫽ confidence interval; RR ⫽ relative risk.
At 268 days, there was a twofold increase in ID revascularization in the angiographic group (7.1% vs 3.5%, p ⫽ 0.02), corresponding to the protocol-scheduled angiogram (Figure 4). This difference primarily resulted from a significant increase in ID target vessel revascularization (nonTLR) in the angiographic group (4.3% vs 1.8%, p ⫽ 0.03). The rate of ID-TLR was numerically higher in the angiographic group than in the clinical follow-up group but the difference was not significant (4.2% vs 2.3%, p ⫽ 0.12). At 3 years, the cumulative incidence of ID revascularization for
angiographic and clinical follow-up groups was similar (Figure 4). Overall TLR (ID and non-ID) was similar between the angiographic and clinical follow-up groups (Table 2), despite a higher rate of non–ID-TLR in the angiographic group (4.5% vs 1.0%, p ⫽ 0.002). Non-TLR target vessel revascularization was similar between the angiographic and clinical follow-up groups for ID and non-ID revascularization. The landmark analysis shows that from the conclusion of the angiographic follow-up window to 3 years, protocol-man-
28
The American Journal of Cardiology (www.ajconline.org)
Figure 6. Ischemia-driven revascularization rates for patients treated with paclitaxel-eluting stents. Abbreviations as in Figure 5.
dated angiographic follow-up compared to clinical follow-up did not affect major adverse cardiovascular events (Figure 2), composite of cardiac death or reinfarction (Figure 3), ID revascularization (Figure 4), cardiac death, myocardial infarction, Academic Research Consortium definite and probable stent thrombosis, TLR, target vessel revascularization (nonTLR), or bleeding complications (Table 3). In aggregate, ID and non–ID-TLR or target vessel revascularization (non-TLR) were similar between groups. However, PES-treated patients assigned to angiographic follow-up did have higher non-ID revascularization rates at 8 months (7.1% vs 0.7%, p ⫽ 0.005) and 3 years (9.7% vs 0.7%, p ⫽ 0.0004) compared to clinical follow-up, and differences were driven by non–ID-TLR rates (6.6% vs 0.7% at 8 months, p ⫽ 0.008, and 9.1% vs 0.7% at 3 years, p ⫽ 0.0007). In contrast, this was not the case in EES-treated patients (Figures 5 and 6). Discussion The principal findings of this subanalysis of the SPIRIT III trial are that compared to clinical follow-up alone, protocol-mandated angiographic follow-up at 8 months ⫾ 28 days (1) did not affect major adverse cardiovascular events or the composite safety end point of cardiac death or myocardial infarction at any time point; (2) was associated with a twofold higher rate of ID revascularization at 8 months but not at 3 years; (3) resulted primarily from a significant increase in ID target vessel revascularization (non-TLR); but (4) increased non–ID-TLR with implantation of PESs but not EESs at 8 months and 3 years.
With the introduction of DESs, the incidence of restenosis has been decreased by up to 60% compared to BMSs.8,11 A decrease in restenosis rates of this magnitude and generally more focal patterns of angiographic restenosis17 are likely to modulate the clinical impact of routine angiographic follow-up in patients treated with DESs. Protocolmandated angiography increased rates of ID revascularization at 9 months; however, by 3 years rates were comparable between the angiographic and clinical follow-up groups, suggesting late catch-up occurs in clinically followed patients. Furthermore, the early increase in ID revascularization was not the result of revascularization of the target lesion treated with the PES or EES device but rather sites remote to the target lesion and therefore had no impact on clinical efficacy measurements of the DESs evaluated in the trial. However, protocol-mandated angiography affected the target lesion by increasing rates of non-ID revascularization after PES but not after EES implantation, a finding that was seen at 9 months and sustained at 3 years. The higher nonischemia TLR rate with PESs may be related to a different visual appearance seen at angiographic follow-up that is not seen with EESs. Lesions treated with PESs (vs EESs) had a significantly larger percent diameter stenosis at 9-month angiographic follow-up (in segment 22.8% vs 18.8%, p ⫽ 0.008; in stent 10.3% vs 5.9%, p ⫽ 0.02), but other measurements including follow-up lesion length and pattern of restenosis were infrequent and not statistically different. These observations differ from comparisons of DESs and BMSs. In an analysis from A Polymer Based, Paclitaxel-
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
Eluting Stent in Patients with Coronary Artery Disease (TAXUS-IV) investigators assessing the impact of protocolmandated angiographic follow-up on the performance of PESs compared to BMSs,17 routine angiographic follow-up resulted in similar increases in rates of target vessel revascularization in patients receiving either stent type. The PES group had a significant relative decrease in target vessel revascularization regardless of angiographic follow-up status; however, angiographic follow-up tended to overestimate the decrease in target vessel revascularization compared to routine clinical follow-up. Our study also demonstrated a systematic twofold increase in ID revascularization at 9 months that was not related, however, to the target site of an EES (p ⫽ 0.32) or a PES (p ⫽ 0.44). In addition, with markedly lower rates of DES restenosis, potential effects at the target lesion are likely to be less apparent in comparisons of DESs. Routine angiographic follow-up provides details not only on the stented segment but also on the entire vascular bed. Therefore, scheduled angiography identifies patients with de novo lesions and in-stent restenosis. The present study suggests this may be less of an issue in the DES era. Similar revascularization rates in the 2 groups at 3 years suggest that the early treatment of these stenoses in the angiographic group is largely offset by revascularization performed later in the clinical follow-up group. Furthermore, early angiography and revascularization do not appear to compromise safety. There does not appear to be an adverse long-term impact in safety in patients undergoing protocol-mandated angiography. At 3 years, rates of cardiac death or myocardial infarction were similar between the angiographic and clinical groups. In the landmark analysis, rates of major adverse cardiovascular events, death, or myocardial infarction were numerically lower with protocol-mandated angiography but not significantly different from the clinical follow-up cohort, providing evidence for lack of a safety concern with systematic angiographic follow-up. The impact of angiographic follow-up on outcomes, overall and by device randomization (EES vs PES), in the SPIRIT III trial are subset analyses and are hypothesis generating. The low event rates associated with DESs may limit the power to detect small differences in subgroups; nevertheless, the absence of meaningful measurable differences in the efficacy measurement of ID-TLR and safety end points provides assurance of the reliability of angiographic surrogate trial designs for DES evaluation. The results of this analysis apply only to the population studied in the SPIRIT III trial. In addition, the different durations of clopidogrel therapy between the angiographic and nonangiographic groups may have influenced major adverse cardiovascular events. 1. Gruntzig A. Transluminal dilatation of coronary-artery stenosis. Lancet 1978;1:263. 2. Gruentzig AR, King SB, Schlumpf M, Siegenthaler W. Long-term follow-up after percutaneous transluminal coronary angioplasty. The early Zurich experience. N Engl J Med 1987;316:1127–1132.
29
3. Park SJ, Kim KH, Oh IY, Shin DH, Park KI, Seo MK, Chung JW, Park KW, Lee HY, Kang HJ, Koo BK, Youn TJ, Kim HS. Comparison of plain balloon and cutting balloon angioplasty for the treatment of restenosis with drug-eluting stents vs bare metal stents. Circ J 2010; 74:1837–1845. 4. Porto I, Burzotta F, Parma A, Pristipino C, Manzoli A, Belloni F, Sardella G, Rigattieri S, Danesi A, Mazzarotto P, Summaria F, Romagnoli E, Prati F, Trani C, Crea F. Angiographic predictors of recurrent stent thrombosis (from the Outcome of PCI for stent-ThrombosIs MultIcentre STudy [OPTIMIST]). Am J Cardiol 2010;105:1710 –1715. 5. Stolker JM, Kennedy KF, Lindsey JB, Marso SP, Pencina MJ, Cutlip DE, Mauri L, Kleiman NS, Cohen DJ; EVENT Investigators. Predicting restenosis of drug-eluting stents placed in real-world clinical practice: derivation and validation of a risk model from the EVENT registry. Circ Cardiovasc Interv 2010;3:327–334. 6. Di Lorenzo E, Sauro R, Varricchio A, Capasso M, Lanzillo T, Manganelli F, Mariello C, Siano F, Pagliuca MR, Stanco G, Rosato G, De Luca G. Benefits of drug-eluting stents as compared to bare metal stent in ST-segment elevation myocardial infarction: four year results of the PaclitAxel or Sirolimus-Eluting stent vs bare metal stent in primary angiOplasty (PASEO) randomized trial. Am Heart J 2009;158(suppl): e43– e50. 7. Mauri L, Silbaugh TS, Garg P, Wolf RE, Zelevinsky K, Lovett A, Varma MR, Zhou Z, Normand SL. Drug-eluting or bare-metal stents for acute myocardial infarction. N Engl J Med 2008;359:1330 –1342. 8. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O’Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE; SIRIUS Investigators. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315–1323. 9. Pandya SB, Kim YH, Meyers SN, Davidson CJ, Flaherty JD, Park DW, Mediratta A, Pieper K, Reyes E, Bonow RO, Park SJ, Beohar N. Drug-eluting versus bare-metal stents in unprotected left main coronary artery stenosis a meta-analysis. JACC Cardiovasc Interv 2010;3: 602– 611. 10. Singh IM, Filby SJ, El Sakr F, Gorodeski EZ, Lincoff AM, Ellis SG, Shishehbor MH. Drug-eluting stents versus bare-metal stents for treatment of bare-metal in-stent restenosis. Catheter Cardiovasc Interv 2010;76:257–262. 11. Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, Turco M, Caputo R, Bergin P, Greenberg J, Popma JJ, Russell ME; TAXUS-IV Investigators. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221– 231. 12. Stone GW, Midei M, Newman W, Sanz M, Hermiller JB, Williams J, Farhat N, Mahaffey KW, Cutlip DE, Fitzgerald PJ, Sood P, Su X, Lansky AJ; SPIRIT III Investigators. Comparison of an everolimuseluting stent and a paclitaxel-eluting stent in patients with coronary artery disease: a randomized trial. jama 2008;299:1903–1913. 13. Chen JL, Yang YJ, Qiao SB, Yao M, Qin XW, Xu B, Liu HB, Wu YJ, Yuan JQ, Chen J, You SJ, Dai J, Li JJ, Gao RL. Comparison of shortand long-term outcomes between Cypher and TAXUS drug-eluting stents for in-stent restenosis. Chin Med Sci J 2007;22:5– 8. 14. Spaulding C. Safety and efficacy update on first-generation drugeluting stents. Am J Cardiol 2008;102(suppl):18J–23J. 15. Cutlip DE, Chauhan MS, Baim DS, Ho KK, Popma JJ, Carrozza JP, Cohen DJ, Kuntz RE. Clinical restenosis after coronary stenting: perspectives from multicenter clinical trials. J Am Coll Cardiol 2002; 40:2082–2089. 16. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW; Academic Research Consortium. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115:2344 –2351. 17. Pinto DS, Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, Mehran R, Na Y, Turco M, Caputo R, Popma JJ, Cutlip DE, Russell ME, Cohen DJ; TAXUS-IV Investigators. Impact of routine angiographic follow-up on the clinical benefits of paclitaxeleluting stents: results from the TAXUS-IV trial. J Am Coll Cardiol 2006;48:32–36.
a
Yale University School of Medicine, New Haven, Connecticut; bKaiser Permanente Medical Center, Los Angeles, California; cAbbott Vascular, Santa Clara, California; dWake Forest University Health Sciences, Winston-Salem, North Carolina; eSt. Vincent’s Heart Center of Indiana, Indianapolis, Indiana; fColumbia University Medical Center, Cardiovascular Research Foundation, New York, New York. Manuscript received July 7, 2011; revised manuscript received and accepted February 21, 2012. This work was funded by Abbott Vascular, Santa Clara, California. Dr. Applegate reports serving on scientific advisory boards and honoraria/ consulting for Abbott Vascular and receiving research funding from Abbott Vascular. Dr. Hermiller reports serving as a consultant for Abbott Vascular and Boston Scientific Corp., Natick, Massachusetts. Dr. Stone reports serving on scientific advisory boards for and honoraria from Abbott Vascular and Boston Scientific Corp. and as a consultant to Medtronic, Santa Rosa, California. *Corresponding author: Tel: 203-737-2142; fax: 203-737-6118. E-mail address: [email protected] (A.J. Lansky). 0002-9149/12/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2012.02.040
remains on the clinical impact of protocol-mandated angiographic follow-up in DES clinical trials, which may preferentially accentuate rates of repeat revascularization and negatively affect safety outcomes.15 Follow-up angiography has been associated with exaggerated rates of revascularization, ostensibly resulting from the “oculostenotic reflex,” a term describing revascularization using percutaneous coronary intervention owing to anatomic lesion severity independent of clinical or physiologic evidence of ischemia. Although this phenomenon is well recognized using balloon angioplasty and BMSs, few investigations have studied the impact of routine angiographic follow-up with DESs. This study aimed to assess the impact of protocol-mandated angiographic follow-up in patients enrolled in A clinical Evaluation of the Investigational Device XIENCE V Everolimus Eluting Coronary Stent System (EECSS) in the Treatment of Subjects With de Novo Native Coronary Artery Lesions (SPIRIT III) trial. Methods The SPIRIT III trial was a prospective, multicenter, randomized, controlled, single-blinded trial in which 1,002 patients were randomized in a 2:1 ratio to the everolimuseluting stent (EES; XIENCE V, Abbott Vascular, Santa www.ajconline.org
22
The American Journal of Cardiology (www.ajconline.org)
Table 1 Baseline characteristics, three-year thienopyridine compliance, and procedural and angiographic characteristics of study population Angiographic Group (n ⫽ 564) Baseline characteristics Demographics Age (years), mean ⫾ SD Men Hypertension Hypercholesterolemia Diabetes mellitus Any Requiring insulin Smoker, current Target vessel Left anterior descending coronary artery Left circumflex coronary artery Right coronary artery Left main coronary artery (protected) Number of narrowed coronary arteries 1 2 ⱖ3 American College of Cardiology/America Heart Association lesion class A B1 B2 C Thienopyridine use At 180 days At 270 days At 365 days At 730 days At 1,095 days Procedural and angiographic characteristics Procedural variables, mean ⫾ SD Number of lesions Number of stents per patient Number of stents per lesion Maximum stent diameter per lesion (mm) Total stent length per lesion (mm) Maximum pressure (atm) Glycoprotein IIb/IIIa inhibitors used
Clara, California) or the paclitaxel-eluting stent (PES; TAXUS Express2, Boston Scientific, Natick, Massachusetts).12 In brief, patients with 1 de novo lesion or 2 de novo coronary lesions with reference vessel diameters of 2.5 to 3.75 mm and lesions ⱕ28 mm in length were eligible for randomization. Telephone randomization was performed in blocks of 3 and 6 patients stratified by diabetes status, 2-vessel treatment, and study site. EESs were available in 2.5-, 3.0-, and 3.5-mm diameters with lengths of 8, 18, and 28 mm. PESs were available in 2.5- to 3.5-mm diameters and lengths ranging from 8 to 32 mm. Operators were instructed to choose a stent length to cover approximately 3 mm of nondiseased tissue at the proximal and distal stent edges. When multiple stents were necessary, 1 to 4 mm of stent overlap was recommended. Operators were not blinded to randomization; however, patients and staff involved in follow-up assessments remained blinded throughout follow-up. The protocol-mandated angiographic fol-
Nonangiographic Group (n ⫽ 438)
p Value
63.3 ⫾ 10.4 387/563 (68.7%) 419/562 (74.6%) 400/554 (72.2%)
62.8 ⫾ 10.4 300/438 (68.5%) 336/438 (76.7%) 323/431 (74.9%)
0.41 0.95 0.46 0.35
160/562 (28.5%) 38/56 (26.8%) 121/551 (22.0%)
29.7% (130/437) 7.3% (32/437) 106/433 (24.5%)
0.67 0.80 0.36
269/647 (41.6%) 182/647 (28.1%) 196/647 (30.3%) 0/647 (0.0%)
212/503 (42.1%) 138/503 (27.4%) 151/503 (30.0%) 2/503 (0.4%)
0.86 0.84 0.95 0.19
378/563 (67.1%) 136/563 (24.2%) 49/563 (8.7%)
276/438 (63.0%) 117/438 (26.7%) 44/438 (10.0%)
0.18 0.38 0.51
44/640 (6.9%) 223/640 (34.8%) 243/640 (38.0%) 130/640 (20.3%)
35/50 (17.0%) 188/501 (37.5%) 199/501 (39.7%) 79/501 (15.8%)
1.00 0.35 0.58 0.05
524/559 (93.7%) 408/549 (74.3%) 372/543 (68.5%) 277/515 (53.8%) 253/512 (49.4%)
411/432 (95.1%) 337/426 (79.1%) 315/417 (75.5%) 263/410 (64.1%) 225/402 (56.0%)
0.41 0.09 0.02 0.002 0.05
647 1.3 ⫾ 0.6 1.2 ⫾ 0.4 3.01 ⫾ 0.37 22.4 ⫾ 8.0 14.8 ⫾ 2.8 150/563 (26.6%)
508 1.3 ⫾ 0.6 1.1 ⫾ 0.4 3.03 ⫾ 0.39 22.3 ⫾ 8.6 15.0 ⫾ 2.8 116/438 (26.5%)
0.38 0.09 0.31 0.89 0.28 1.00
low-up was specified at 8 months for the first 564 randomized patients in the study. A landmark analysis was performed from the last day of the 8th-month angiographic follow-up window (268 days) or from 9 months to 3 years. Clinical end points used in the SPIRIT III trial have been previously defined.12 In brief, major adverse cardiovascular events were defined as the composite of cardiac death, myocardial infarction, and ischemia-driven target lesion revascularization (ID-TLR). Myocardial infarction was defined as the development of new pathologic Q waves ⱖ0.4 second in duration in ⱖ2 contiguous leads or as an increase in creatinine phosphokinase-MB. Revascularization was categorized as ID or non- ID. ID revascularization required (1) a positive functional study, or (2) the presence of symptoms and a ⬎50% diameter stenosis by independent core laboratory, or (3) in the absence of symptoms or a positive functional study required ⬎70% diameter stenosis by an independent core
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
23
Table 2 Outcomes of entire cohort at three years Result/Outcome Cardiac death Myocardial infarction Q wave Non-Q wave Stent thrombosis (Academic Research Consortium) Definite Probable Definite or probable Target lesion revascularization, all Ischemia driven Nonischemia driven Target vessel revascularization (nontarget lesion revascularization), all Ischemia driven Nonischemia driven Bleeding complications
Angiographic Group (n ⫽ 564)
Nonangiographic Group (n ⫽ 438)
p Value
9/522 (1.7%) 21/522 (4.0%) 2/522 (0.4%) 19/522 (3.6%)
7/412 (1.7%) 23/412 (5.6%) 3/412 (0.7%) 20/412 (4.9%)
1.00 0.28 0.66 0.41
4/523 (0.8%) 2/523 (0.4%) 6/523 (1.1%) 55/532 (10.3%) 37/532 (7.0%) 24/532 (4.5%) 49/532 (9.2%) 43/532 (8.1%) 7/532 (1.3%) 23/515 (4.5%)
4/408 (1.0%) 3/408 (0.7%) 7/408 (1.7%) 31/416 (7.5%) 27/416 (6.5%) 4/416 (1.0%) 27/416 (6.5%) 26/416 (6.3%) 3/416 (0.7%) 25/405 (6.2%)
0.74 0.66 0.58 0.14 0.80 0.002 0.15 0.31 0.53 0.30
Figure 1. Patient flow in the SPIRIT III randomized clinical trial through 3 years depicting split of patients treated with everolimus-eluting or paclitaxeleluting stents into an angiographic versus a nonangiographic group.
laboratory. ID revascularization was further adjudicated by the angiographic core laboratory based on the location of recurrence as ID-TLR and ID target vessel revascularization (non-TLR or revascularization remote from the original target site). Non-ID-TLR was defined as revascularization in the absence of any of the previously described criteria. Protocol-defined stent thrombosis was (1) an acute coronary syndrome with angiographic evidence of thrombus within or adjacent to a previously treated lesion, or (2), in the absence of angiographic evidence, any unexplained death or ST-segment elevation myocardial infarction, or (3) new Q waves in the distribution of the target lesion within 30 days of the index percutaneous coronary intervention. Stent thrombosis was also
independently adjudicated according to the Academic Research Consortium definition. All clinical events were adjudicated by an independent clinical events committee (Harvard Clinical Research Institute).16 All baseline, follow-up, and event-related angiograms were analyzed by an independent angiographic core laboratory (Cardiovascular Research Foundation, New York, New York) by analysts blinded to treatment allocation using standard definitions and methods for all quantitative measurements using Medis software (Medis, Leiden, the Netherlands). All outcomes are reported at 3 years from the index percutaneous coronary intervention and the 9-month (268 days, upper window of the 8th month) landmark date and are based on the intent-to-treat principle.
24
The American Journal of Cardiology (www.ajconline.org)
Figure 2. Cumulative incidence of major adverse cardiovascular events (MACEs) in patients undergoing routine angiographic follow-up and those not scheduled for angiographic follow-up. (A) Time from index percutaneous coronary intervention to 3 years. (B) Nine-month/268-day (last day of 8th month window) landmark to 3 years. Hazard ratios (HRs) and log-rank p values are calculated at 268 and 1,123 days, respectively.
The cohort was divided into patients undergoing protocol-mandated angiography at 8 months and those undergoing standard clinical follow-up. During the index percutaneous coronary intervention, each patient was treated with an EES or a PES. Landmark survival analysis was used to determine whether there were differences in the primary outcome between routine angiographic and standard clinical follow-up groups after 268 days (last day of 8th month angiographic follow-up window) or the 9-month landmark. Patients in the standard clinical follow-up group who died, developed myocardial infarction, or underwent repeat revascularization within 268 days
of the index percutaneous coronary intervention were excluded from the landmark analysis. Similarly, patients in the routine angiographic group who died, developed myocardial infarction, or underwent repeat revascularization before the scheduled angiographic follow-up were excluded from the landmark analysis. Continuous variables are reported as mean ⫾ SD and were compared by t test. Categorical variables are reported as absolute values and percentages with comparisons made with the Fisher exact test. The Kaplan–Meier method was used to construct survival curves for the primary end point of a major adverse cardiovascular event (cardiac death,
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
25
Figure 3. Cumulative incidence of cardiac death and myocardial infarction (MI) in patients undergoing routine angiographic follow-up and those not scheduled for angiographic follow-up. (A) Time from index percutaneous coronary intervention to 3 years. (B) Nine-month/268-day (last day of 8th month window) landmark to 3 years. Hazard ratios and log-rank p values are calculated at 268 and 1,123 days, respectively. Other abbreviation as in Figure 2.
myocardial infarction, or ID-TLR) and its components. Comparisons between groups were made with log-rank test. Cox proportional hazards models were used to generate hazard ratios and 95% confidence intervals for the study outcomes. The proportionality assumption of the proportional hazards model was assessed and satisfied using log(⫺log) plots. All tests are 2-tailed, with differences reported as statistically significant if the p value was ⬍0.05. All analyses were performed with SAS 9.1 (SAS Institute, Cary, North Carolina).
Results The SPIRIT III trial enrolled 1,002 patients at 65 United States sites from June 2005 through March 2006. Patients were randomized to receive EESs (n ⫽ 669) or PESs (n ⫽ 333). In total 564 randomized patients were assigned to undergo coronary angiography at 8 months. Of these, 435 (77%) underwent angiography as planned at 240 ⫾ 28 days or had a previous angiogram that qualified for analysis. Angiographic and clinical follow-up groups were well
26
The American Journal of Cardiology (www.ajconline.org)
Figure 4. Cumulative incidence of ischemia-driven revascularization, composite of ischemia-driven target lesion revascularization, and ischemia-driven target vessel revascularization. (A) Time from index percutaneous coronary intervention to 3 years. (B) Nine-month landmark to 3 years. Hazard ratios and log-rank p values are calculated at 268 and 1,123 days respectively. Other abbreviation as in Figure 2.
matched for cardiac risk factors, location, and severity of coronary artery disease (Table 1) and for procedural and angiographic characteristics (Table 1). All patients received a thienopyridine at discharge (Table 1), with a shorter mean duration of therapy in the angiographic compared to the clinical follow-up group (682 ⫾ 398 vs 750 ⫾ 376 days, p ⫽ 0.006, respectively). Thienopyridine use was similar at 6 months (180 days) between groups (Table 1). At 9 months, 1 year, 2 years, and 3 years, clopidogrel use was greater in the clinically followed group. In contrast, aspirin use was similar between the angiographic and clinical follow-up groups at all time points, with a mean duration of aspirin
therapy of 999 ⫾ 241 days in the angiographic group and 997 ⫾ 253 days in the clinical follow-up group (p ⫽ 0.92). Follow-up at 3 years was available in 934 patients (93.2% of study cohort), with similar follow-up rates between groups (Table 2, Figure 1). There was no significant difference in 3-year major adverse cardiovascular events between the angiographic and clinical follow-up groups (Figure 2), and the composite of cardiac death and myocardial infarction was also similar (Figure 3). There were no differences between the 2 groups in myocardial infarction, Academic Research Consortium definite or probable stent thrombosis, or bleeding complications (Table 2).
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
27
Table 3 Nine-month (268-day) landmark analysis outcomes at three years Result/Outcome Cardiac death Myocardial infarction Q wave Non-Q wave Stent thrombosis (Academic Research Consortium) Definite Probable Definite or probable Target lesion revascularization, all Ischemia driven Nonischemia driven Target vessel revascularization (nontarget lesion revascularization), all Ischemia driven Nonischemia driven Bleeding complications
Angiographic Group (n ⫽ 564)
Nonangiographic Group (n ⫽ 438)
p Value
5/462 (1.1%) 9/462 (1.9%) 1/462 (0.2%) 8/462 (1.7%)
6/386 (1.6%) 12/386 (3.1%) 3/386 (0.8%) 9/386 (2.3%)
0.56 0.38 0.34 0.63
1/516 (0.2%) 1/516 (0.2%) 2/516 (0.4%) 17/470 (3.6%) 13/470 (2.8%) 5/470 (1.1%) 19/470 (4.0%) 17/470 (3.6%) 2/470 (0.4%) 9/459 (2.0%)
3/406 (0.7%) 3/406 (0.7%) 6/406 (1.5%) 19/390 (4.9%) 17/390 (4.4%) 2/390 (0.5%) 16/390 (4.1%) 16/390 (4.1%) 2/390 (0.5%) 5/381 (1.3%)
0.33 0.33 0.15 0.40 0.26 0.47 1.00 0.72 1.00 0.59
Figure 5. Ischemia-driven revascularization rates for patients treated with everolimus-eluting stents. CI ⫽ confidence interval; RR ⫽ relative risk.
At 268 days, there was a twofold increase in ID revascularization in the angiographic group (7.1% vs 3.5%, p ⫽ 0.02), corresponding to the protocol-scheduled angiogram (Figure 4). This difference primarily resulted from a significant increase in ID target vessel revascularization (nonTLR) in the angiographic group (4.3% vs 1.8%, p ⫽ 0.03). The rate of ID-TLR was numerically higher in the angiographic group than in the clinical follow-up group but the difference was not significant (4.2% vs 2.3%, p ⫽ 0.12). At 3 years, the cumulative incidence of ID revascularization for
angiographic and clinical follow-up groups was similar (Figure 4). Overall TLR (ID and non-ID) was similar between the angiographic and clinical follow-up groups (Table 2), despite a higher rate of non–ID-TLR in the angiographic group (4.5% vs 1.0%, p ⫽ 0.002). Non-TLR target vessel revascularization was similar between the angiographic and clinical follow-up groups for ID and non-ID revascularization. The landmark analysis shows that from the conclusion of the angiographic follow-up window to 3 years, protocol-man-
28
The American Journal of Cardiology (www.ajconline.org)
Figure 6. Ischemia-driven revascularization rates for patients treated with paclitaxel-eluting stents. Abbreviations as in Figure 5.
dated angiographic follow-up compared to clinical follow-up did not affect major adverse cardiovascular events (Figure 2), composite of cardiac death or reinfarction (Figure 3), ID revascularization (Figure 4), cardiac death, myocardial infarction, Academic Research Consortium definite and probable stent thrombosis, TLR, target vessel revascularization (nonTLR), or bleeding complications (Table 3). In aggregate, ID and non–ID-TLR or target vessel revascularization (non-TLR) were similar between groups. However, PES-treated patients assigned to angiographic follow-up did have higher non-ID revascularization rates at 8 months (7.1% vs 0.7%, p ⫽ 0.005) and 3 years (9.7% vs 0.7%, p ⫽ 0.0004) compared to clinical follow-up, and differences were driven by non–ID-TLR rates (6.6% vs 0.7% at 8 months, p ⫽ 0.008, and 9.1% vs 0.7% at 3 years, p ⫽ 0.0007). In contrast, this was not the case in EES-treated patients (Figures 5 and 6). Discussion The principal findings of this subanalysis of the SPIRIT III trial are that compared to clinical follow-up alone, protocol-mandated angiographic follow-up at 8 months ⫾ 28 days (1) did not affect major adverse cardiovascular events or the composite safety end point of cardiac death or myocardial infarction at any time point; (2) was associated with a twofold higher rate of ID revascularization at 8 months but not at 3 years; (3) resulted primarily from a significant increase in ID target vessel revascularization (non-TLR); but (4) increased non–ID-TLR with implantation of PESs but not EESs at 8 months and 3 years.
With the introduction of DESs, the incidence of restenosis has been decreased by up to 60% compared to BMSs.8,11 A decrease in restenosis rates of this magnitude and generally more focal patterns of angiographic restenosis17 are likely to modulate the clinical impact of routine angiographic follow-up in patients treated with DESs. Protocolmandated angiography increased rates of ID revascularization at 9 months; however, by 3 years rates were comparable between the angiographic and clinical follow-up groups, suggesting late catch-up occurs in clinically followed patients. Furthermore, the early increase in ID revascularization was not the result of revascularization of the target lesion treated with the PES or EES device but rather sites remote to the target lesion and therefore had no impact on clinical efficacy measurements of the DESs evaluated in the trial. However, protocol-mandated angiography affected the target lesion by increasing rates of non-ID revascularization after PES but not after EES implantation, a finding that was seen at 9 months and sustained at 3 years. The higher nonischemia TLR rate with PESs may be related to a different visual appearance seen at angiographic follow-up that is not seen with EESs. Lesions treated with PESs (vs EESs) had a significantly larger percent diameter stenosis at 9-month angiographic follow-up (in segment 22.8% vs 18.8%, p ⫽ 0.008; in stent 10.3% vs 5.9%, p ⫽ 0.02), but other measurements including follow-up lesion length and pattern of restenosis were infrequent and not statistically different. These observations differ from comparisons of DESs and BMSs. In an analysis from A Polymer Based, Paclitaxel-
Coronary Artery Disease/Angiographic Follow-up After PCI Using DES
Eluting Stent in Patients with Coronary Artery Disease (TAXUS-IV) investigators assessing the impact of protocolmandated angiographic follow-up on the performance of PESs compared to BMSs,17 routine angiographic follow-up resulted in similar increases in rates of target vessel revascularization in patients receiving either stent type. The PES group had a significant relative decrease in target vessel revascularization regardless of angiographic follow-up status; however, angiographic follow-up tended to overestimate the decrease in target vessel revascularization compared to routine clinical follow-up. Our study also demonstrated a systematic twofold increase in ID revascularization at 9 months that was not related, however, to the target site of an EES (p ⫽ 0.32) or a PES (p ⫽ 0.44). In addition, with markedly lower rates of DES restenosis, potential effects at the target lesion are likely to be less apparent in comparisons of DESs. Routine angiographic follow-up provides details not only on the stented segment but also on the entire vascular bed. Therefore, scheduled angiography identifies patients with de novo lesions and in-stent restenosis. The present study suggests this may be less of an issue in the DES era. Similar revascularization rates in the 2 groups at 3 years suggest that the early treatment of these stenoses in the angiographic group is largely offset by revascularization performed later in the clinical follow-up group. Furthermore, early angiography and revascularization do not appear to compromise safety. There does not appear to be an adverse long-term impact in safety in patients undergoing protocol-mandated angiography. At 3 years, rates of cardiac death or myocardial infarction were similar between the angiographic and clinical groups. In the landmark analysis, rates of major adverse cardiovascular events, death, or myocardial infarction were numerically lower with protocol-mandated angiography but not significantly different from the clinical follow-up cohort, providing evidence for lack of a safety concern with systematic angiographic follow-up. The impact of angiographic follow-up on outcomes, overall and by device randomization (EES vs PES), in the SPIRIT III trial are subset analyses and are hypothesis generating. The low event rates associated with DESs may limit the power to detect small differences in subgroups; nevertheless, the absence of meaningful measurable differences in the efficacy measurement of ID-TLR and safety end points provides assurance of the reliability of angiographic surrogate trial designs for DES evaluation. The results of this analysis apply only to the population studied in the SPIRIT III trial. In addition, the different durations of clopidogrel therapy between the angiographic and nonangiographic groups may have influenced major adverse cardiovascular events. 1. Gruntzig A. Transluminal dilatation of coronary-artery stenosis. Lancet 1978;1:263. 2. Gruentzig AR, King SB, Schlumpf M, Siegenthaler W. Long-term follow-up after percutaneous transluminal coronary angioplasty. The early Zurich experience. N Engl J Med 1987;316:1127–1132.
29
3. Park SJ, Kim KH, Oh IY, Shin DH, Park KI, Seo MK, Chung JW, Park KW, Lee HY, Kang HJ, Koo BK, Youn TJ, Kim HS. Comparison of plain balloon and cutting balloon angioplasty for the treatment of restenosis with drug-eluting stents vs bare metal stents. Circ J 2010; 74:1837–1845. 4. Porto I, Burzotta F, Parma A, Pristipino C, Manzoli A, Belloni F, Sardella G, Rigattieri S, Danesi A, Mazzarotto P, Summaria F, Romagnoli E, Prati F, Trani C, Crea F. Angiographic predictors of recurrent stent thrombosis (from the Outcome of PCI for stent-ThrombosIs MultIcentre STudy [OPTIMIST]). Am J Cardiol 2010;105:1710 –1715. 5. Stolker JM, Kennedy KF, Lindsey JB, Marso SP, Pencina MJ, Cutlip DE, Mauri L, Kleiman NS, Cohen DJ; EVENT Investigators. Predicting restenosis of drug-eluting stents placed in real-world clinical practice: derivation and validation of a risk model from the EVENT registry. Circ Cardiovasc Interv 2010;3:327–334. 6. Di Lorenzo E, Sauro R, Varricchio A, Capasso M, Lanzillo T, Manganelli F, Mariello C, Siano F, Pagliuca MR, Stanco G, Rosato G, De Luca G. Benefits of drug-eluting stents as compared to bare metal stent in ST-segment elevation myocardial infarction: four year results of the PaclitAxel or Sirolimus-Eluting stent vs bare metal stent in primary angiOplasty (PASEO) randomized trial. Am Heart J 2009;158(suppl): e43– e50. 7. Mauri L, Silbaugh TS, Garg P, Wolf RE, Zelevinsky K, Lovett A, Varma MR, Zhou Z, Normand SL. Drug-eluting or bare-metal stents for acute myocardial infarction. N Engl J Med 2008;359:1330 –1342. 8. Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O’Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE; SIRIUS Investigators. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349:1315–1323. 9. Pandya SB, Kim YH, Meyers SN, Davidson CJ, Flaherty JD, Park DW, Mediratta A, Pieper K, Reyes E, Bonow RO, Park SJ, Beohar N. Drug-eluting versus bare-metal stents in unprotected left main coronary artery stenosis a meta-analysis. JACC Cardiovasc Interv 2010;3: 602– 611. 10. Singh IM, Filby SJ, El Sakr F, Gorodeski EZ, Lincoff AM, Ellis SG, Shishehbor MH. Drug-eluting stents versus bare-metal stents for treatment of bare-metal in-stent restenosis. Catheter Cardiovasc Interv 2010;76:257–262. 11. Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, Turco M, Caputo R, Bergin P, Greenberg J, Popma JJ, Russell ME; TAXUS-IV Investigators. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med 2004;350:221– 231. 12. Stone GW, Midei M, Newman W, Sanz M, Hermiller JB, Williams J, Farhat N, Mahaffey KW, Cutlip DE, Fitzgerald PJ, Sood P, Su X, Lansky AJ; SPIRIT III Investigators. Comparison of an everolimuseluting stent and a paclitaxel-eluting stent in patients with coronary artery disease: a randomized trial. jama 2008;299:1903–1913. 13. Chen JL, Yang YJ, Qiao SB, Yao M, Qin XW, Xu B, Liu HB, Wu YJ, Yuan JQ, Chen J, You SJ, Dai J, Li JJ, Gao RL. Comparison of shortand long-term outcomes between Cypher and TAXUS drug-eluting stents for in-stent restenosis. Chin Med Sci J 2007;22:5– 8. 14. Spaulding C. Safety and efficacy update on first-generation drugeluting stents. Am J Cardiol 2008;102(suppl):18J–23J. 15. Cutlip DE, Chauhan MS, Baim DS, Ho KK, Popma JJ, Carrozza JP, Cohen DJ, Kuntz RE. Clinical restenosis after coronary stenting: perspectives from multicenter clinical trials. J Am Coll Cardiol 2002; 40:2082–2089. 16. Cutlip DE, Windecker S, Mehran R, Boam A, Cohen DJ, van Es GA, Steg PG, Morel MA, Mauri L, Vranckx P, McFadden E, Lansky A, Hamon M, Krucoff MW, Serruys PW; Academic Research Consortium. Clinical end points in coronary stent trials: a case for standardized definitions. Circulation 2007;115:2344 –2351. 17. Pinto DS, Stone GW, Ellis SG, Cox DA, Hermiller J, O’Shaughnessy C, Mann JT, Mehran R, Na Y, Turco M, Caputo R, Popma JJ, Cutlip DE, Russell ME, Cohen DJ; TAXUS-IV Investigators. Impact of routine angiographic follow-up on the clinical benefits of paclitaxeleluting stents: results from the TAXUS-IV trial. J Am Coll Cardiol 2006;48:32–36.