YEAR IN CARDIOLOGY SERIES
The Year in Interventional Cardiology
Simon R. Dixon, MBChB, FRACP, FACC*,*,
Cindy L. Grines, MD, FACC* and
William W. O'Neill, MD, FACC
* William Beaumont Hospital, Royal Oak, Michigan
University of Miami, Miami, Florida.
Manuscript received January 24, 2008;
revised manuscript received March 17, 2008,
accepted March 18, 2008.
* Reprint requests and correspondence: Dr. Simon R. Dixon, Division of Cardiology, William Beaumont Hospital, 3601 West 13 Mile Road, Royal Oak, Michigan 48073. (Email: sdixon{at}beaumont.edu).
We are honored to provide readers of the Journal with this review of major scientific work published in the field of Interventional Cardiology in 2007. In addition, we have included late-breaking trials presented at the American College of Cardiology (ACC), Transcatheter Cardiovascular Therapeutics, and American Heart Association (AHA) conferences. We hope that the paper will provide a broad overview of the field for general cardiologists, as well as a framework for more detailed study for those with a specific interest in Interventional Cardiology.
 |
ACCF/AHA/Society for Cardiovascular Angiography and Interventions 2007 Clinical Competence Statement
|
|---|
An update of the 1998 clinical competence statement on cardiac interventional procedures was published in 2007 (1). This important document provides recommendations for assessment and maintenance of cognitive knowledge and technical skills necessary for proficiency in cardiac interventional procedures. The statement has also been expanded to include percutaneous noncoronary interventions, including aortic and mitral valvuloplasty, atrial septal defect and patent foramen ovale closure, and alcohol septal ablation. The document reviews expected success and complication rates, comorbidities or other risk factors that may be used for risk adjustment, and the relationships between operator and institutional activity and percutaneous coronary intervention (PCI) outcomes. A few key recommendations with respect to operator volumes include: 1) for PCI, the operator volume threshold continues to be 75 procedures/year; 2) for atrial septal defect/patent foramen ovale closure, a minimum of 10 cases to gain clinical competence, with a further 10 procedures/year to maintain competency; 3) for alcohol septal ablation, a minimum of 10 initial procedures, and 6 cases/year thereafter; and 4) for valvuloplasty, a minimum or 5 to 10 cases with an experienced colleague before attempting valvuloplasty independently. For further details, readers are encouraged to review the Statement (1).
|
Acute Myocardial Infarction (MI)
|
|---|
Facilitated PCI.
There has been ongoing interest in the pharmacoinvasive approach to improve outcomes of mechanical reperfusion. A small study randomized 41 patients to receive intracoronary streptokinase (250 ku) versus none after primary PCI (2). Two days after PCI, the streptokinase group had better measures of microvascular function and improved Thrombolysis In Myocardial Infarction (TIMI) frame count; however, at 6 months there was no difference in left ventricular (LV) size or function.
Maioli et al. (3) randomized 210 ST-segment elevation myocardial infarction (STEMI) patients to receive early abciximab (in the emergency center) versus late administration (in cath lab). Patients treated with early abciximab showed benefits in pre-PCI TIMI flow grade 3 (24% vs. 10%; p = 0.01), better frame counts, and blush scores. Moreover, post-PCI ST-segment resolution (50% vs. 35% of patients; p = 0.3) and 1-month improvement in ejection fraction (EF) (8% vs. 6%; p = 0.2) favored the early abciximab group. In contrast, the larger FINESSE (Facilitated Intervention With Enhanced Reperfusion Speed to Stop Events) trial found no benefit of up-front abciximab before PCI (4). The FINESSE trial enrolled 2,452 patients within 4 h of STEMI and randomized them to 1 of 3 arms: abciximab-facilitated PCI, half-dose retaplase with abciximab-facilitated PCI, or primary PCI. At 90 days, there was no difference in the primary composite end point or individual end points of death, ventricular fibrillation, cardiogenic shock, or readmission for congestive heart failure. However, bleeding complications were increased in the facilitated arms.
Rescue PCI.
Another meta-analysis of rescue PCI including 8 randomized trials was published in 2007 (5). This confirmed a previous meta-analysis (6) showing that rescue PCI-treated patients had a trend for lower mortality (7.3% vs. 10.4%; p = 0.09) and significant reduction in congestive heart failure (12.7% vs. 17.8%; p = 0.05) and reinfarction (6.1% vs. 10.7%; p = 0.04). However, rescue PCI was associated with an increased risk of ischemic stroke and minor bleeding compared with conservative care.
PCI after thrombolysis.
The CARESS in AMI (Combined Abciximab Retaplase Stent Study in Acute Myocardial Infarction) trial (7) treated 600 STEMI patients <75 years in age with half-dose reteplase, abciximab, aspirin, and heparin, then randomized to immediate transfer for PCI versus transfer only for rescue PCI (persistent ST-segment elevation). The combined end point of death, MI, or refractory ischemia was improved in the immediate PCI group (4.4% vs. 10.7%; p = 0.004). The rate of TIMI major or minor bleeding was higher in the immediate PCI group, but this was not statistically significant. These data support previous trials showing that immediate PCI after thrombolysis is beneficial (compared with conservative care), but did not investigate the role of primary PCI (which was the superior strategy in FINESSE [4] and in the meta-analysis of 17 facilitated PCI trials) (8).
In the ExTRACT–TIMI 25 (Enoxaparin and Thrombolysis Reperfusion for Acute Myocardial Infarction Treatment) trial, 4,676 patients underwent PCI after fibrinolytic therapy (9). Compared to unfractionated heparin, enoxaparin was associated with a lower incidence of death or MI (10.7% vs. 13.8%; p = 0.001), with no difference in major bleeding.
Time-to-treatment.
Brodie et al. used the EMERALD (Enhanced Myocardial Efficacy and Removal by Aspiration of Liberated Debris) trial to determine the impact of time-to-treatment on final infarct size assessed by sestamibi imaging (10). Reperfusion <2 h after symptom onset was associated with a smaller infarct size (2% vs. 11%; p = 0.008); however, only 5% of patients were reperfused this quickly. After 3 h, there was little change in final infarct size with increasing delay. Time-to-treatment was more important for anterior versus inferior MI. These findings, along with data from Nallamothu et al. (11) showing that only 35% of U.S. patients have door-to-balloon time of  90 min, suggest that major changes must occur in the U.S. hospital systems to improve access and speed to reperfusion. Khot et al. (12) described how a change in hospital policy (emergency department activation of the cath lab) decreased treatment delay by 30 min and led to a marked increase from 28% to 75% in the number of patients being treated within 90 min.
Outcomes research.
A review of the Solucient database demonstrated a decline in the rate of hospitalizations for AMI from 2002 to 2005 (309 to 226 per 100,000 persons), driven primarily by a decrease in transmural MI admissions (13).
Previous studies have demonstrated higher mortality for women admitted with AMI. Milcent et al. (14) reviewed 74,389 hospitalizations (30% women) in France in 1999. Women were older and had more comorbidities, including valve disease, heart failure, stroke, and diabetes. Crude mortality was higher in women (14.8% vs. 6.1%; p < 0.0001). After adjustment for age and comorbidities, women still had a higher mortality.
Similarly, racial differences in the use of revascularization and mortality were reported (15). Popescu et al. (15) studied 1.2 million Medicare beneficiaries admitted with AMI from January 2000 to June 2005. Rates of revascularization and transfer from non-PCI to PCI facilities were lower for black versus white patients. Although in-hospital mortality rates were similar, long-term mortality was higher for black patients (15).
Several groups reported on development of regional systems for STEMI care. In the RACE (Reperfusion of Acute Myocardial Infarction in North Carolina Emergency Departments) study, marked improvements in time-to-reperfusion were observed at both PCI and non-PCI centers after implementation of a statewide reperfusion program (16). Henry et al. (17) and Ting et al. (18) described 2 regional systems for transfer of STEMI patients in Minnesota. These studies demonstrated that rural hospitals can triage and successfully treat STEMI patients with mechanical reperfusion with well organized transfer protocols.
Ross et al. (19) reviewed the Medicare database for 137,279 AMI patients admitted from 1994 to 1996 to determine the impact of the certificate-of-need (CON) program. These authors found that states with CON statutes used revascularization less frequently when equivocal indications were present.
Adjunctive therapies and devices.
Results of several trials evaluating pharmacologic adjuncts to mechanical reperfusion were reported in 2007. In the APEX-MI (Assessment of Pexelizumab in Acute Myocardial Infarction) trial, 5,745 patients undergoing primary PCI were randomized to receive intravenous pexelizumab or placebo. There was no difference in the incidence of 30-day mortality (4.06% for pexelizumab vs. 3.92% for placebo) (20). Díaz et al. (21) reported findings of the OASIS-6 GIK (Organization to Assess Strategies for Ischemic Syndromes 6–Glucose-Insulin-Potassium) trial, as well as a combined analysis of these data with the CREATE-ECLA (Clinical Trial of Metabolic Modulation in Acute Myocardial Infarction Treatment Evaluation–Estudios Cardiologicos Latinoamerica). Overall, infusion of glucose-insulin-potassium had no benefit on survival or other clinical outcomes. On a positive note, the J-WIND (Japan Working Groups of Acute Myocardial Infarction for the Reduction of Necrotic Damage by Atrial Natriuretic Peptide or Nicorandil) investigators found that administration of intravenous atrial natriuretic peptide in patients with AMI resulted in a 14% reduction in infarct size assessed by creatine kinase release kinetics (22).
The AMIHOT (Acute Myocardial Infarction with Hyperoxemic Therapy) trials investigated the use of intracoronary supersaturated oxygen therapy in patients with AMI (23,24). Patients with anterior MI <6 h treated with supersaturated oxygen had a 25% reduction in infarct size compared with the control group (median infarct size 18.5% vs. 25% of the left ventricle; p = 0.023).
The DELTA MI (Direct Inhibition of  Protein Kinase C Enyzme to Limit Total Infarct Size in Acute Myocardial Infarction) trial tested 4 escalating doses of KAI-9803 in patients with anterior wall MI and total occlusion of the left anterior descending artery (25). Favorable improvements in ST-segment resolution, creatine kinase release kinetics, myocardial perfusion grade, and infarct size were observed.
Recent studies have suggested there is no role for routine use of thrombectomy or embolic protection in AMI. In the PREMIAR (Protection from Distal Embolization in High-Risk Patients With Acute ST-Segment Elevation Myocardial Infarction) trial, use of the SpideRx filter (ev3, Plymouth, Minnesota) during PCI had no benefit on the extent of ST-segment resolution, myocardial blush, EF, or clinical outcomes (26). Similar results were found in a larger study of 626 patients using the FilterWire device (27). De Luca et al. published a meta-analysis of 21 trials of devices designed to prevent distal embolization during mechanical reperfusion (28). Among 3,721 patients, use of adjunctive devices was associated with better myocardial perfusion and less embolization; however, there was no impact on survival. One important issue with many of these earlier studies has been defining which patients have significant thrombus burden, particularly when the vessel is totally occluded. In this regard, a recent report by Sianos et al. (29) has been insightful. Thrombus burden was assessed in 812 STEMI patients before and after flow was established with either a guidewire or small balloon. Of note, only 28% of patients had a large thrombus burden after reclassification, whereas 67% patients would have been classified in this group if baseline angiography alone were used to define thrombus (29).
Drug-eluting stents.
Results of several studies were published in 2007. The SESAMI (Sirolimus-Eluting Stent Versus Bare-Metal Stent in Acute Myocardial Infarction) trial randomized 320 STEMI patients to implantation of either a sirolimus-eluting stent (SES) or a bare-metal stent (BMS) (30). At 1 year, the incidence of binary restenosis (the primary end point) was lower in the SES group (9.3% vs. 21.3%; p = 0.032), as was the incidence of TLR and MACE. Valgimigli et al. provided follow-up data from the STRATEGY (Single High-Dose Bolus Tirofiban and Sirolimus Eluting Stent Versus Abciximab) trial and found the initial benefit of SES implantation was still present at 2 years (31). However, a report of 505 STEMI patients from the RESEARCH (Rapamycin-Eluting Stent Evaluated at Rotterdam Cardiology Hospital) and T-SEARCH (Taxus-Stent Evaluated at Rotterdam Cardiology Hospital) registries concluded that use of either a SES or a paclitaxel-eluting stent (PES) was not associated with a lower rate of major adverse cardiac events (MACE) at 3-year follow-up (32). Finally, Kastrati et al. (33) published a meta-analysis of 8 randomized trials comparing drug-eluting stents (DES) with BMS in 2,786 AMI patients. The DES significantly reduced the need for reintervention (hazard ratio 0.38, 95% CI 0.29 to 0.50; p < 0.001). Importantly, there was no difference in the incidence of stent thrombosis (ST), MI, or death between patients receiving DES or BMS.
A randomized trial comparing a titanium-nitric oxide-coated stent versus PES in 425 AMI patients was reported (34). At 1 year, the primary end point MACE was similar in both arms (10.3% vs. 12.8%; p = 0.5), but there was a lower incidence of ST in the titanium-nitric oxide stent group.
Cell-based cardiac repair.
The REPAIR AMI (Reinfusion of Enriched Progenitor Cells and Infarct Remodeling) study demonstrated that intracoronary bone marrow cell (BMC) infusion improved EF. A Doppler substudy in 58 of 204 patients found that vessels treated with BMC infusion had a 90% increase in coronary flow reserve, resulting in normalization by 4 months (p = 0.005 compared with placebo) (35), suggesting that BMC infusion induces healing of damaged endothelium and preservation of microvascular function.
Leone et al. (36) randomized 41 patients with large anterior wall MI (EF <50%) to granulocyte-specific colony-stimulating factor (G-CSF) injections for 5 days (n = 14) versus control (n = 27). In this small series, G-CSF prevented LV dilatation and improved EF (36). Conversely, other studies have found no benefit of G-CSF (37). To investigate this lack of benefit, Ripa et al. (38) quantified and characterized the types of cells mobilized by G-CSF in the STEMMI (Stem Cells in Myocardial Infarction) trial. They found a lower ratio of potentially beneficial mesenchymal stem cells to the number of leukocytes and an inverse relationship between mesenchymal stem cells and EF.
Cardiogenic shock.
Systemic inflammation, including generation of excess nitric oxide, is believed to contribute to the pathogenesis of persistent cardiogenic shock in AMI patients. The TRIUMPH (Tilarginine Acetate Injection in a Randomized lnternational Study in Unstable MI Patients With Cardiogenic Shock) trial was designed to examine the effect of nitric oxide synthase inhibition with tilarginine acetate on 30-day mortality; however, the study was terminated after enrollment of 398 patients (planned 658 patients) based on a pre-specified futility analysis (mortality 48% tilarginine, 42% placebo) (39). In another study, admission hemoglobin was found to be an independent predictor of 1-year mortality in shock patients (40).
|
Acute Coronary Syndromes
|
|---|
The ACC/AHA guidelines for acute coronary syndromes (ACS) were updated in 2007 (41). Notable changes include more emphasis on use of troponin as the dominant cardiac biomarker, earlier clinical evaluation to reduce treatment delay, B-type natriuretic peptide to supplement risk assessment, use of multislice computerized tomography angiography and cardiac magnetic resonance imaging, addition of fondaparinux and bivalirudin as alternatives to unfractionated heparin, and more support for thienopyridine use, including upstream use and longer administration. A consensus has emerged that early PCI use results in favorable outcomes, especially in high-risk patients. On the other hand, results of the ICTUS (Invasive Versus Conservative Treatment in Unstable Coronary Syndromes) trial suggest that an initial conservative strategy with "watchful waiting" may be considered in stabilized low-risk patients (42). An AHA scientific statement discussed the value of PCI in the elderly (greater absolute and relative benefit with PCI) (43). Unfortunately, the elderly also have a greater likelihood of bleeding complications. Insight into U.S. practice patterns was provided from the CRUSADE (Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA Guidelines) registry on 82,924 patients with non-STEMI treated from 2002 to 2005 (44). During this time, PCI use increased from 38% to 52% of cases. In addition, PCI use in patients with 3-vessel disease increased by 9% and coronary artery bypass graft (CABG) use decreased by the same extent. The clinical sequelae of bleeding after PCI were highlighted in a substudy of the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) trial (45). These authors found that patients with major hemorrhage had higher 30-day mortality (7.3% vs. 1.2%), and rates of ischemia and stent thrombosis were dramatically increased in these patients. Bivalirudin use decreased the risk of major hemorrhage from 5.7% to 3.0% (p < 0.001).
|
Elective PCI
|
|---|
PCI versus medical therapy.
Rarely has one clinical study generated so much controversy, or had such a marked and immediate impact on clinical practice, as the COURAGE (Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation) trial. Boden et al. (46) screened 36,000 patients with mild stable angina and randomized 2,287 (6.3%) to PCI or medical therapy (47). The primary end point of death or MI was the same for both groups. Angina was significantly less at follow-up in the PCI group. This trial corroborates previous guidelines that suggest PCI be used only for symptomatic relief in this population. Limitations of the trial, including lack of power and high expected event rate were described by Kereiakes et al. (48). The findings of the COURAGE trial were largely confirmed in the 5-year follow-up of the MASS (Medicine, Angioplasty, or Surgery Study) trial in which patients randomized to surgery, PCI, or medical therapy had similar outcomes for PCI or medical therapy (49).
Perhaps a more powerful observation concerning treatment of silent ischemia after MI was seen in the SWISSI II (Swiss Interventional Study on Silent Ischemia Type II) trial. Erne et al. (50) randomized 201 patients with recent MI and stress-induced ischemia to PCI or medical therapy. Percutaneous coronary intervention resulted in lower MACE and better preservation of ventricular function.
PCI versus CABG.
The final 10-year report of the BARI (Bypass Angioplasty Revascularization Investigation) trial was published (51). Overall, no difference in mortality was observed between PCI and CABG patients. Patients with treated diabetes lived longer after CABG (58% vs. 46%; p = 0.025). Moreover, in a retrospective registry, diabetic patients with 2- and 3-vessel disease who underwent PCI with a DES had a lower survival at 1 year compared with bypass patients (52). This study further swings the pendulum toward surgery for multivessel diabetic patients.
Outcomes research.
In 2007, there were several reports addressing clinical outcomes after PCI. Singh et al. (53) provided long-term outcome data on 24,410 PCI procedures performed at the Mayo Clinic over 25 years. Significant reductions in in-hospital mortality and need for emergency bypass surgery were observed over time. In addition, long-term risk-adjusted clinical outcomes, including survival, were significantly improved over the 25-year period.
Kansagra et al. (54) studied trends in operator and hospital volume and outcomes in 452,404 patients undergoing PCI in Florida and New York from 1996 to 2001. The risk-adjusted end point of mortality and emergency CABG decreased over the time period from 2.8% to 1.6%, with the lowest incidence of death/CABG observed with operators performing  75 PCI procedures at hospitals with >400 procedures.
Anderson et al. (55) provided a report on the usefulness of the American College of Cardiology-National Cardiovascular Data Registry (ACC-NCDR) risk-adjustment model to predict in-hospital mortality across the 4 classes of PCI indication (classes I, IIa, IIb, and III). The incidence of observed major cardiovascular complications increased from class I to class III. Estimated mortalities for each class using the ACC-NCDR risk model were close to observed values and were similar to estimates derived using the Michigan and New York State risk models.
Feldman et al. (56) reported outcome data from 47,020 patients in the New York State Angioplasty Registry comparing those who underwent ad hoc versus staged PCI. The incidence of in-hospital mortality, MACE, and renal failure was similar in both groups, suggesting the strategy of ad hoc PCI is safe in contemporary practice.
Although primary PCI may be performed at hospitals without cardiac surgical backup, there has been controversy about performing elective PCI at hospitals without on-site surgery. Dehmer et al. (57) examined the frequency of PCI at facilities without on-site surgery in the ACC-NCDR. From 2001 to 2005, there was a marked increase from 3.2% to 16% in the number of centers performing PCI without backup. Of note, there was a significant increase in the number of elective PCI procedures, as well as nonelective cases, at facilities without surgical backup.
Recently, there has been debate about the appropriateness of specialty care cardiac hospitals. In a study of Medicare beneficiaries, Nallamothu et al. (58) compared the population-based rates of coronary revascularization after opening of a cardiac hospital. Hospital referral regions (HRR) with opening of a cardiac hospital had a 2-fold higher number of revascularization procedures 4 years after opening, compared with HRRs where new cardiac programs opened at a general hospital or with HRRs with no new program.
In other reports, baseline anemia was found to be an independent predictor of mortality and adverse clinical outcomes after PCI (59,60).
Same-day discharge.
Heyde et al. (61) evaluated the safety of same-day discharge in 800 patients undergoing elective PCI via the femoral approach (5- or 6-F). Four hours after PCI, patients were triaged as suitable for early discharge or not ( 80% were considered to be stable for early discharge). Overall, the primary end point (death, MI, CABG, repeat PCI, or puncture-related complications) occurred in 2.2% of same-day discharge patients and 4.2% of overnight-stay patients, suggesting that the early discharge strategy may be safe in carefully selected patients.
Bifurcation and intermediate lesions.
Percutaneous treatment of bifurcation lesions remains challenging. In a multicenter registry, Tsuchikane et al. (62) used directional coronary atherectomy before DES implantation and reported a low rate of angiographic restenosis at 9 months (main branch 1.1%, side-branch 3.4%) and MACE. Grube et al. (63) reported results of a novel self-expanding drug-eluting bifurcation stent (Axxess Plus, Devax, Irvine, California). Procedural success with the stent was >90%, and in-stent late loss was 0.09 mm.
Pijls et al. (64) presented 5-year clinical follow-up of the DEFER (Deferral Versus Performance of Percutaneous Coronary Angioplasty in Patients Without Documented Ischemia) trial. Among patients with a fractional flow reserve (FFR) of 0.75, clinical outcome was excellent, with no difference in event-free survival between those who were assigned to deferral or performance of PCI (80% vs. 73%; p = 0.52). Readers requiring more in-depth study on assessment of intermediate coronary lesions with FFR and intravascular ultrasound (IVUS) should refer to a comprehensive review published this past year (65).
Distal embolization.
Several articles published in 2007 investigated the pathophysiology of distal embolization during PCI. Bahrmann et al. (66) reported on use of an intracoronary Doppler guidewire for real-time detection of coronary embolization. Patients with periprocedural myonecrosis had a higher frequency of microembolization during PCI. Kawamoto et al. (67) extended these observations and found that the risk of embolization was related to the presence of necrotic core assessed with virtual histology intravascular ultrasound. Among patients with STEMI, Kawaguchi et al. (68) demonstrated that the volume of the necrotic core was significantly related to the risk of ST-segment re-elevation during primary PCI. Selvanayagam et al. (69) used cardiac magnetic resonance imaging to assess the effect of embolization on myocardial perfusion reserve index and found that 21 out of 40 patients demonstrated new distal hyperenhancement after PCI.
|
DES
|
|---|
Late stent thrombosis.
Over the past year, there have been an enormous number of publications comparing DES with BMS, confirming the benefits of reduction in restenosis and TVR but also calling attention to a small increased risk of late ST after DES implantation. Several scientific reports, review articles, and editorials have described the scope of the problem (70–77). A comprehensive review of these studies (78) concluded that: 1) ST is an infrequent but very severe complication of both BMS and DES; 2) during 4 years of follow-up from randomized trials comparing DES and BMS, there is no difference in overall ST frequency, although the time course differs, with an excess of ST late after DES implant; 3) despite this relative imbalance, no differences in the end points of death or death/MI between DES and BMS are observed; and 4) longer-term follow-up of these patients as well as larger datasets are required to fully study this issue.
Mechanisms of DES Thrombosis
Finn et al. (79) reported data from 81 human autopsies of DES, of which 28 lesions were found to have ST. The most powerful predictor of ST was lack of endothelialization, particularly the ratio of uncovered struts to total struts. Similarly, Awata et al. (80) compared serial angioscopic findings from 17 patients treated with SES with 11 patients treated with BMS. Neointimal coverage was typically complete by 3 to 6 months after BMS; however, SES-treated vessels had thrombi and yellow plaques visualized even 2 years after implantation.
The exact mechanism of delayed endothelialization is unknown. Cook et al. (81) reported IVUS data from 13 patients with DES thrombosis compared with 144 patients who had routine IVUS follow-up but no ST. Stent thrombosis occurred more frequently with long lesions, multiple stents, or overlapping stents. Although the size of the reference vessel was nearly identical between groups, in patients with ST the cross-sectional area of the external elastic membrane was significantly larger at the stent site, suggesting that positive remodeling may have led to incomplete stent apposition. Another contributing factor to DES thrombosis may be abnormal vasomotor function of the coronary. Obata et al. (82) studied 25 patients with AMI due to left anterior descending artery occlusion treated with SES (n = 13) or BMS (n = 16). Two weeks after AMI, the response to acetylcholine was diminished in vessels treated with SES, suggesting endothelium-dependent vasomotor dysfunction.
Prevention of DES Thrombosis
Two additional studies in 2007 noted the strong association of DES thrombosis with discontinuation of thienopyridine therapy (83,84). A scientific advisory committee was convened to make recommendations for dual antiplatelet therapy after DES (85). A minimum of 12 months' dual antiplatelet therapy was recommended for any patient receiving a DES (Table 1), and there was consensus agreement to postpone elective surgery for at least 1 year after DES (86). Dual antiplatelet therapy beyond 1 year is recommended by some, especially if patient and lesion characteristics predict DES thrombosis. Eisenstein et al. (87) confirmed the potential benefits of prolonged clopidogrel use after DES, with reduction in the risk of death or MI out to 24 months. In addition to high clopidogrel loading doses (600 mg), one may consider a higher maintenance dose (150 mg) in diabetics and patients who may be clopidogrel resistant (135).
Clinical outcomes after DES.
Randomized Trials
Several meta-analyses (88–93) have described long-term clinical outcomes from randomized trials of DES (Table 2). All reports demonstrate no difference in long-term mortality or overall ST between DES and BMS. Stettler et al. (92) conducted a meta-analysis of all 38 trials (18,023 patients) which prospectively randomized patients to DES, including trials of primary PCI and off-label indications. At 4-year follow-up, mortality and the risk of ST was similar between DES and BMS. Interestingly, SES were associated with a significant reduction in the risk of MI compared with BMS (hazard ratio [HR] 0.81, 95% confidence interval [CI] 0.66 to 0.97; p = 0.03). Moreover, SES seemed to be clinically better than PES in reducing target lesion revascularization (TLR) (HR 0.70; p = 0.0021), MI (HR 0.83; p = 0.045), and late ST (p = 0.041).
Registries
Several large multicenter or national registries comparing DES with BMS have confirmed the safety of DES (94–102). Although the original SCAAR (Swedish Coronary Angiography and Angioplasty Registry) publication suggested an increase in late events in DES-treated patients (98), a larger more updated report showed a 50% reduction in restenosis and similar long-term mortality compared with BMS (99). Furthermore, there was a significant reduction in MI/death within the first 6 months after DES. Similarly, registries from Canada, Denmark, Italy, and the U.S. have demonstrated that use of DES was associated with reduced target vessel revascularization (TVR) with either superior or similar rates of death and MI compared with BMS (Table 3).
Off-label use of DES.
Beohar et al. (103) reported a multicenter registry of 5,541 patients receiving DES, of which 47% were for off-label or untested indications. The 30-day risk of death, MI, or ST and 1-year rate of TVR was significantly higher when DES was used off-label compared with approved indications; however, absolute event rates were quite low. Win et al. (104) reported a multicenter registry of 3,323 patients treated with DES, of whom 55% had at least 1 off-label indication. They noted similar mortality but a higher risk of MI, ST, and TVR in off-label compared with FDA-approved DES indications. These "off-label" patients are a high-risk population with numerous comorbidities, unfavorable lesion morphology, and unstable clinical presentations, and are expected to have less favorable outcomes after either BMS or DES. However, large real-world registries including off-label applications generally demonstrated that DES was associated with reduced TVR with either superior or similar rates of death and MI compared with BMS (Table 3).
Colombo and Chieffo (105) provided an excellent review of publications addressing off-label use of DES. Based on observational and case-controlled series as well as small randomized trials, they concluded that in some lesion subsets (chronic occlusions, bifurcations, small vessels, long lesions, vein grafts) the data appear to support the use of DES. They caution against the routine use of DES in triple-vessel disease, diabetics with multivessel disease, AMI, and unprotected left main coronary artery until more data and longer follow-up are available.
Unprotected Left Main Disease
Chieffo et al. (106) reported a multicenter registry in which 147 patients received DES for unprotected left main coronary artery (ULMCA) lesions (distal bifurcation lesions were excluded). Procedural success was 99%, and at long-term follow-up (886 ± 308 days), rates of death (3.4%) and TVR (4.7%) were low. The Sheiban et al. (107) registry of 85 patients with ULMCA treated with SES noted that 61% of lesions involved the distal bifurcation (43% received single stent, 57% modified T or V stenting). At 2-year follow-up, death occurred in only 2.4%, MI in 3.6%, and TVR in 10.8%.
Sanmartin et al. (108) retrospectively compared 96 patients with ULMCA treated with DES with 245 patients treated with bypass surgery. Surgically treated patients had a trend for higher combined death, stroke, and Q-wave MI during long-term follow-up (p = 0.07). However, the surgical group was higher risk and had longer follow-up. Moreover, TVR was higher in the DES group.
Erglis et al. (109) conducted a randomized trial of BMS versus PES in 103 patients with ULMCA. All PCIs were IVUS guided and used cutting balloon followed by stenting. Although 75% of lesions were in the distal LMCA, the investigators used a single stent strategy jailing the circumflex. At 6 months, restenosis occurred in 22% BMS and 6% PES patients (p = 0.021), and MACE in 30% and 13%, respectively (p = 0.054).
Therefore, DES for ULMCA may be a reasonable approach in selected patients, but routine application is premature. An editorial reviewing the use of stents for ULMCA addressed numerous unanswered questions and the need to await results of the SYNTAX (Synergy Between Percutaneous Coronary Intervention With TAXUS DES and Cardiac Surgery) trial (110).
Lesions At High Risk of Restenosis
Baumgart et al. (111) reported 1-year results in 200 diabetics randomized to SES versus BMS. In-segment restenosis was significantly reduced with SES (8.8% vs. 42.1%; p < 0.0001), as were TLR (5.3% vs. 21.1%; p = 0.002) and MACE (14.7% vs. 35.8%).
At 18 months' follow-up of a prospective randomized trial, DES were found to be cost-effective in lesions at high risk of restenosis (vessel size <3.0 mm or saphenous vein graft lesions) (112). In a subgroup analysis of the SIRTAX (Sirolimus-Eluting Stent Compared With Paclitaxel-Eluting Stent for Coronary Revascularization) randomized trial, Togni et al. (113) confirmed the benefit (reduced MACE and TVR) of SES over PES in small vessels. Similarly, Petronio et al. (114) found significantly reduced neointimal hyperplasia assessed by IVUS in a randomized trial of SES versus PES in long left anterior descending artery lesions.
Dibra (115) performed a meta-analysis of 4 trials, which randomized 1,230 patients with restenosis to DES versus percutaneous transluminal coronary angioplasty or brachytherapy. Drug-eluting stents were associated with significant reductions in TLR (odds ratio [OR] 0.35, 95% CI 0.25 to 0.49; p < 0.001) and angiographic restenosis (OR 0.36, 95% CI 0.27 to 0.49; p = 0.001).
On a more somber note, the only randomized trial of DES in vein grafts showed initial benefit, but long-term follow-up at 32 months demonstrated loss of the initial reduction in TLR, as well as higher mortality (29% vs. 0%; p = 0.001) (116). However, given the small sample size (n = 75), mortality differences may be due to chance.
DES restenosis.
Drug-eluting stents have greatly reduced, but not eliminated, restenosis. When DES restenosis is managed percutaneously, it is associated with a high rate of MACE (117). Oral medications may further reduce the incidence of DES restenosis. One trial randomized 274 patients treated with PES to receive celecoxib (400 mg before PCI and 200 mg twice daily for 6 months) versus control (118), and found reduced late loss (0.49 mm vs. 0.75 mm) and TLR (5% vs. 15%; p = 0.008) with celecoxib. A larger study randomized 500 patients with long lesions (32 mm) requiring DES to cilostazol (200 mg load, 100 mg twice daily for 6 months) versus control (119). The cilostazol group had reduced late loss (0.34 mm vs. 0.51 mm; p < 0.001) and TLR (2.8% vs. 6.8%; p = 0.036) in this subgroup of patients at high risk of DES restenosis.
New DES.
Daemen and Serruys (120) provided an outstanding review of current and future DES. They discussed new stent platforms, including magnesium and bioabsorbable stents, new coatings and polymers to reduce inflammatory reactions, heparin coatings combined with SES, and biodegradable polymers. Finally, they reviewed the drug coatings, including new paclitaxel stents, 6 different limus-related antiproliferative drugs, combinations of limus drugs with dexamethasone or paclitaxel, and the prohealing approach using antibodies targeted at endothelial progenitor cells.
Publications of new DES were limited in 2007. Turco et al. (121) reported results from a prospective registry using the Taxus Liberté stent platform (designed to be more deliverable). Despite applications to more complex lesions, clinical outcomes and late loss were equivalent to an earlier cohort of patients treated with the Taxus Express platform. Erbel et al. (122) reported the use of 71 bioabsorbable magnesium stents and found early success, but late loss at 4 months was high (1.08 mm), and by 1 year TVR was 45%. Neointimal growth and negative remodeling were the mechanisms of restenosis.
Adriaenssens et al. (123) randomized 502 patients to a rapamycin-eluting stent (RES) plus estradiol versus RES alone (both stents were polymer free). At 1 year, late loss and clinical events were similar between the 2 groups. However, the study did not assess whether estrogen promoted rapid re-endothelialization (which is the potential benefit of estradiol) and was not powered to determine differences in ST.
|
Brachytherapy
|
|---|
Ferrero et al. (124) reported 5-year follow-up of the BetAce trial of brachytherapy for treatment of de novo lesions. Although brachytherapy reduced the need for TVR at 6 months, this benefit was not sustained. Lee et al. (125) compared outcomes after SES versus brachytherapy for diffuse in-stent restenosis. At 3 years, the incidence of TLR and MACE was significantly lower in patients treated with SES compared with brachytherapy.
|
Saphenous Vein Graft Disease
|
|---|
The PROXIMAL (Proximal Protection During Saphenous Vein Graft Intervention) trial evaluated use of proximal embolic protection with the Proxis Embolic Protection System (St. Jude Medical, Maple Grove, Minneapolis) (126). A total of 594 patients were randomized in a noninferiority design to either control (distal protection whenever possible) or test (proximal protection when possible, distal when not). The 30-day MACE rate was 9.2% in the test arm and 10.0% in the control arm, suggesting that proximal embolic protection provides similar outcomes to distal protection devices.
Results of the AMEthyst (Assessment of the Medtronic AVE Interceptor Saphenous Vein Graft Filter System) trial, which evaluated the Medtronic Interceptor Plus distal filter (Medtronic, Minneapolis, Minnesota), were presented this past year (127). At 30 days, the primary end point MACE occurred in 8.0% of patients in the Interceptor arm and 7.2% of patients in the control device arm (GuardWire or FilterWire) (p = 0.027 for noninferiority), suggesting that this novel filter provides a similar degree of protection to the other approved devices.
Despite the benefit of embolic protection, recent data from the ACC-NCDR are sobering (128). Among 19,546 patients undergoing vein graft intervention, an embolic protection device was used in only 22%. Use of an embolic protection device was associated with a lower incidence of no-reflow (OR 0.68; p = 0.032) but no difference in in-hospital mortality. The universal occurrence of distal embolization in vein grafts was highlighted in a report from van Gaal et al. (129) in which 57 out of 58 patients were found to have visible debris after PCI. Moreover, embolization was not predicted by baseline clinical or angiographic variables, thus underscoring the importance of embolic protection in vein graft intervention.
|
Pharmacotherapy
|
|---|
Clopidogrel.
Inadequate response to antiplatelet therapy was nicely summarized in a review article (130). Mahmud et al. (131) investigated 150 patients treated with eptifibatide and found inadequate platelet inhibition (PI) in 61% of patients after a single bolus, and in 36% after a double bolus. A fibrinogen level >75 mg/dl was a strong predictor of inadequate PI.
Bliden et al. (132) studied 100 patients treated with aspirin and clopidogrel for more than 1 month before PCI and found a strong association between post-PCI ischemic events and high platelet aggregation. Similarly, Buonamici et al. (133) found that 13% of 804 patients treated with 600 mg clopidogrel had inadequate PI, and those patients suffered an 8.6% incidence of ST within 6 months. Campo et al. (134) performed platelet aggregation studies in 143 patients taking clopidogrel, and then repeated studies after crossing over to ticlopidine. Inadequate PI to either clopidogrel or ticlopidine at steady state was common, but nonresponse to both drugs was infrequent (3.5%), suggesting that inadequate PI is drug specific rather than class specific. Angiolillo et al. (135) randomized 40 patients with type 2 diabetes mellitus (a disease known to have inadequate PI) to clopidogrel maintenance doses of 75 mg versus 150 mg/day. Although the high-dose group had significant (p = 0.002) improvement in PI, 60% of diabetics still had suboptimal clopidogrel response.
Prasugrel, a new thienopyridine with greater and more consistent PI, was compared with clopidogrel in 13,608 ACS patients undergoing PCI (136). The primary end point of cardiovascular death, MI, or stroke was improved with prasugrel (9.9% vs. 12.1%; p < 0.001), as was MI (7.4% vs. 9.7%; p < 0.001), urgent TVR (2.5% vs. 3.7%; p < 0.001), and ST (1.1% vs. 2.4%; p < 0.001). However, life-threatening bleeding was increased (1.4% vs. 0.9%; p = 0.01), and patients with previous stroke had net clinical harm from prasugrel.
Glycoprotein IIb/IIIa receptor inhibitor.
To determine whether the duration of glycoprotein IIb/IIIa inhibitor infusion could be shortened after uncomplicated PCI, the BRIEF-PCI (Brief Infusion of Eptifibatide Following PCI) study randomized 624 patients (45% ACS, 70% clopidogrel pre-treatment) to either a 2-h or a standard 18-h infusion of eptifibatide (137). The primary end point, incidence of ischemic injury, was 30.1% in the short group and 28.3% in the standard group. Major bleeding was less frequent in the short group. These data suggest that the infusion of eptifibatide can be abbreviated after successful PCI.
The EVA-AMI (Eptifibatide Versus Abciximab in Primary PCI for Acute ST Elevation Myocardial Infarction) study randomized 429 STEMI patients undergoing primary PCI to receive either abciximab or eptifibatide (138). There was no significant difference in the primary end point, the extent of ST-segment resolution 60 min after PCI.
Antithrombin therapy.
Stone et al. (139) compared use of bivalirudin and unfractionated heparin or enoxaparin for patients undergoing PCI. Outcomes, summarized in Table 4, suggest no difference in ischemic end points but a major decline in risk of bleeding events for patients treated with bivalirudin. Of note, however, subgroup analysis showed a small increase in ischemic events in patients who were not pre-treated with clopidogrel. In a planned subgroup analysis of the ACUITY trial, Stone described compared upstream versus intraprocedural glycoprotein IIb/IIIa use (140). Upstream use did not decrease ischemic events (7.7% vs. 4.1%; p = NS), but increased major bleeding (6.1% vs. 4.9%; p < 0.001). Feit et al. (141) further highlighted the importance of bleeding complications in a subanalysis of the REPLACE (Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events) 2 trial. In this report, 3.2% of patients had major hemorrhage after PCI, and 1-year mortality was significantly higher for patients with major bleeding (8.7% vs. 1.9%; p < 0.001).
Other new anticoagulation regimens during PCI were tested. Mehta et al. (142) analyzed 6,238 PCI patients treated with fondaparinux versus enoxaparin in the OASIS-5 trial. Major bleeding was reduced by fondaparinux use (2.4% vs. 5.1%; p < 0.00001); however, catheter thrombosis was more frequent with fondaparinux (0.9% vs. 0.4%). Another new direct factor Xa inhibitor, otamixaban, was tested in 947 patients in the SEPIA-PCI (Study to Evaluate the Pharmacodynamics, the Safety and Tolerability, and the Pharmacokinetics of Several Intravenous Regimens of the Factor Xa Inhibitor Otamixaban (XRP0673), in Comparison to Intravenous Unfractionated Heparin in Subjects Undergoing Non-Urgent Percutaneous Coronary Intervention) trial (143). The agent had efficacy at higher doses, but catheter thrombosis was also reported in patients treated at low dosages without IIb/IIIa inhibitors. Both the OASIS and the SEPIA-PCI trials highlight a major, pitfall of factor Xa inhibitor use during PCI.
Finally, Teirstein's group reported a novel anticoagulation regimen in the REMOVE trial (144). These investigators used a triple antiplatelet therapy (aspirin, clopidogrel, eptifibatide) and randomized patients to unfractionated heparin versus no heparin. Bleeding (as assessed by the Landefeld bleeding index) was reduced in the no-heparin group (3.0% vs. 3.7%; p = 0.03). Owing to the small sample size, larger confirmatory trials of this approach will be required.
Statins.
Pre-treatment with high-dose statins appear to have a protective effect in patients undergoing PCI. The ARMYDA-ACS (Atorvastatin for Reduction of Myocardial Damage During Angioplasty) trial randomly assigned 171 patients with a non–ST-segment elevation ACS to pre-treatment with atorvastatin (80 mg 12 h before PCI) or placebo (145). In this small trial, patients in the atorvastatin arm had a significantly lower incidence of 30-day MACE (5% vs. 17%; p = 0.01), driven mostly by a lower incidence of MI within the first 24 h.
|
Contrast Nephropathy
|
|---|
Prevention of contrast-induced nephropathy (CIN) remains an important focus of ongoing research. The CARE (Cardiac Angiography in Renally Impaired Patients) trial compared the effect of iopamidol (low osmolar) and iodixanol (iso-osmolar) in 482 patients with baseline creatinine clearance <60 ml/min (146). There was no difference in the incidence of CIN between the low-osmolar and iso-osmolar agents (4.4% vs. 6.7%; p = 0.39).
In 2007, there were 4 studies investigating the effect of sodium bicarbonate infusion for prevention of CIN (147–150) (Table 5). In the largest trial, bicarbonate infusion had no effect on the incidence of CIN (147). Three other studies, however, demonstrated a lower incidence of CIN with bicarbonate (148–150). Until further data are forthcoming, it seems reasonable to use bicarbonate infusion in patients at risk for CIN, given the ease of this approach and lack of alternative protective therapies.
View this table:
[in this window]
[in a new window]
|
Table 5 Trials Evaluating Use of Sodium Bicarbonate Infusion for Prevention of Contrast-Induced Nephropathy in 2007
|
|
For patients with more severe chronic renal failure, prophylactic hemodialysis appears to be a useful strategy. Lee et al. (151) randomized 82 patients with serum creatinine >3.5 mg/dl to hemodialysis (started as soon as possible after angiography, duration 4 h) or normal saline alone. Patients treated with prophylactic hemodialysis had a smaller change in renal function at 72 h and a significantly lower need for temporary or permanent renal replacement therapy.
Laskey et al. (152) reported a novel technique to predict the risk of an increase in serum creatinine after PCI based on the ratio of contrast media to creatine clearance (V/CrCl). In 3,179 patients undergoing PCI, a V/CrCl ratio >3.7 was identified as a significant independent predictor of post-PCI creatinine increase (OR 3.84, 95% CI 2.0 to 7.3; p < 0.001). This useful index can be applied to estimate the maximum amount of contrast for a patient, above which there is a likelihood of developing acute deterioration of renal function.
|
Structural Heart Disease
|
|---|
Percutaneous aortic valve replacement.
Further clinical trial experience with the Edwards pericardial stent valve was reported by Webb et al. (153) and Walther et al. (154). Webb reported the Vancouver experience on 50 patients treated via the transfemoral approach. The 30-day mortality was 12%. No structural deterioration in valve function was observed at 1 year. Walther et al. (154) reported the surgical multicenter experience in 59 patients treated by a transapical approach. The in-hospital mortality was 13.6%. These data serve as prelude to pivotal U.S. FDA approval trials with this technology.
Patent foramen ovale and atrial septal defect.
Handke et al. (155) provided a report suggesting that patent foramen ovale (PFO) has a strong association with cryptogenic stroke in both young and old patients. For patients <55 years old with cryptogenic stroke, 44% had a PFO versus 14% in patients with a known cause of stroke. For patients >55 years old, PFO was also more frequent (28% vs. 12%; p < 0.001). These data continue to fuel interest in the concept that PFO closure may decrease risk of embolic stroke.
Devices for transcatheter closure of PFO continue to be evaluated. Jones et al. (156) reported the multicenter U.S. experience in 119 patients with atrial septal defect treated with the Helex septal occluder device. Radiofrequency thermal coaptation of PFO was tested in a porcine model by Hara et al. (157). Sievert et al. (158) tested another radiofrequency device in 30 patients.
Atrial appendage and other devices.
Sick et al. (159) reported use of the Watchman left atrial appendage occluder system (Atritech, Plymouth, Minnesota). Sixty-six patients who were candidates for systemic anticoagulation were treated and followed-up for 740 days. There were 2 nondevice-related deaths, 2 patients with tamponade after implantation, and 2 patients had a transient ischemic attack at follow-up. Banai et al. (160) reported implantation of a coronary sinus reducer to treat refractory angina. Six out of 9 patients had an improvement in angina and exercise-induced ischemia.
|
Peripheral Vascular Disease
|
|---|
Carotid disease.
In 2007, a Clinical Expert Consensus Document on carotid artery stenting was published (161). The document includes recommendations on evaluation, diagnostic imaging, medical therapy, carotid revascularization, and credentialing. Van der Heyden et al. (162) investigated the use of carotid stenting before planned cardiac surgery in patients with severe asymptomatic carotid disease. Procedural success was 97.7%, and there was a low incidence of death and stroke from time of carotid stent to 30 days after surgery (4.8%), suggesting that this is a valuable alternative treatment option in patients with combined carotid and cardiac disease. Kao et al. (163) provided a report on the feasibility of endovascular recanalization for symptomatic internal carotid artery occlusion in 30 patients. Technical success was 73%. Finally, Goldstein (164) published a comprehensive review on the treatment of acute ischemic stroke.
Peripheral disease.
The optimal treatment of superficial femoral artery (SFA) disease has been controversial. Krankenberg et al. (165) published results of the FAST (Femoral Artery Stenting Trial), in which 244 patients with SFA disease and chronic limb ischemia were randomized to implantation of a single nitinol stent or balloon angioplasty alone. At 1 year, the primary end point of ultrasound-assessed binary restenosis was not statistically different between the treatment arms (38.1% for angioplasty vs. 31.7% for stent; p = 0.37), nor was there any difference in TLR. Finally, Mahmud et al. (166) published an excellent review article on the current role of endovascular therapy in the management of subclavian, carotid, renal, aortoiliac, and infrainguinal disease.
|
Footnotes
|
|---|
Dr. Dixon is a consultant for Boston Scientific, ev3, and TherOx. Dr. Grines is a consultant for Pfizer, Sanofi-Synthelabo, Bristol-Myers Squibb, and GlaxoSmithKline Global Cardiovascular Board. Dr. O'Neill is a consultant for Guidant, Medtronic, Johnson & Johnson, Cordis, and Boston Scientific and owns equity in TherOx, Radiant Medical, and Accumed. William Beaumont Hospital Division of Cardiology has received research support from Abbott, Abiomed, AstraZeneca, Aventis, Boston Scientific, Cardiac Assist, Cordis, Esperion, ev3, GlaxoSmithKline, InnerCool, Medtronic, Otsuka, Pfizer, Radiant Medical, and TherOx.
|
References
|
|---|
|
Top
ACCF/AHA/Society for...
Acute Myocardial Infarction (MI)