Important Interventional Trials From AHA.19

Introduction

The American Heart Association's Scientific Sessions 2019 provided a significant number of trials impacting interventional cardiology. These trials include the long-awaited ISCHEMIA (International Study of Comparative Health Effectiveness With Medical and Invasive Approaches) and two trials, TWILIGHT (Ticagrelor With Aspirin or Alone in High-Risk Patients After Coronary Intervention) and AUGUSTUS (Antithrombotic Therapy After Acute Coronary Syndrome or PCI in Atrial Fibrillation) focusing on antiplatelet therapy post-stent in routine and special populations. Others describe special aspects of coronary management: COMPLETE (Complete Versus Culprit-Only Revascularization Strategies to Treat Multivessel Disease After Early PCI for STEMI) describes the timing of complete revascularization during ST-segment elevation myocardial infarction (STEMI), and another trial assesses outcomes for percutaneous ventricular assist devices versus intra-aortic balloon pumps (IABP). A transcatheter aortic valve implantation outcome registry is also presented.

Collected here are excellent summaries of and commentary on these trials by members of the American College of Cardiology Interventional Council. I thank them for their excellent efforts. I trust you, our readers, will find these summaries informational and useful to position the applicability of the described results.

ISCHEMIA (International Study of Comparative Health Effectiveness With Medical and Invasive Approaches)

By George W. Vetrovec, MD, MACC
Editorial Team Lead, Invasive Cardiovascular Angiography & Interventions collection on ACC.org
Richmond, VA

ISCHEMIA addressed whether revascularization plus optimal medical treatment was associated with better outcomes than medications alone in patients with moderate to severe ischemia. Moderate to severe ischemia was documented by a variety of stress modalities but ultimately did not use imaging in all cases, relying instead on ST-segment analysis. Patient exclusions included those with left ventricular ejection fraction <35%, recent myocardial infarction (MI), severely limiting angina, and/or New York Heart Failure Class III or IV. To exclude significant left main disease, all patients were blindly screened with computed tomography angiography to exclude >50% obstruction. Overall, 5,179 patients were randomized, including 77% male patients and 41% patients with diabetes.

At a median follow-up of 3.3 years with the composite endpoint of cardiovascular death, MI, hospitalization for unstable angina, hospitalization for heart failure (HF), or resuscitated cardiac arrest, there was no difference in the 2 treatment groups (invasive group = 13.3%; medical group = 15.5%; p = 0.34). Importantly, the results were consistent across all endpoints because a late (pre-specified, if needed) expansion of the composite endpoint did not affect the results. Overall mortality was low (approximately 6.5%) as were other endpoints, suggesting favorable benefits of medical management, which in the trial was adequate but certainly not optimal. Separate analyses showed a symptomatic benefit for revascularization persistent over the follow-up period. A separate study of renal insufficiency patients showed no benefit from revascularization in this population.

Additional unanswered scientific questions remain. Is there truly equipoise between the early MI group and the late MI group? If the early MI group are those with small, periprocedural enzyme elevations below a level impacting late outcomes, then there may be a potentially significant difference. An in-depth analysis of the current MI endpoints is promised at the coming American College of Cardiology Scientific Sessions 2020 Together With World Congress of Cardiology. Most importantly, long-term follow-up is critical to follow this important endpoint. Hopefully the study leaders will be successful in obtaining funding for this important follow-up.

How do we best identify left main disease and severe left ventricular dysfunction and reaffirm to practitioners that not all angina should be treated medically based on the exclusions in ISCHEMIA?

The most glaring limitation is whether, in fact, the results demonstrate that eliminating ischemia is unimportant. Although there was no difference in outcome in patients undergoing percutaneous coronary intervention (PCI), we have no data on how competently and completely revascularization was achieved (i.e., to what extent ischemia was eliminated to warrant this conclusion). There are strong, albeit mostly observational, data showing better outcomes for complete revascularization for PCI and coronary artery bypass grafting (particularly in comparisons with off-pump surgery). Presenter Dr. Hochman acknowledged at the meeting that no data were currently available regarding the extent of revascularization and likely the level of ischemia relief achieved. Until these data are available, it seems that the basic scientific question concerning the impact of ischemia relief on long-term outcome in the management of moderately ischemic coronary artery disease. As prior observational studies would certainly suggest, there is potentially a difference in outcome dependent on the extent of ischemia relief.

Until these questions are answered, we must be certain that the management of ischemic disease is carefully individualized. We can all agree that not every bump in a coronary artery requires a stent, that there remain subgroups of potential benefit, and that we should not treat all angina with pills without appropriate diagnostic evaluation.

TWILIGHT (Ticagrelor With Aspirin or Alone in High-Risk Patients After Coronary Intervention) ACS Substudy

By Mirvat A. Alasnag, MD
King Fahd Armed Forces Hospital
Jeddah, Saudi Arabia

Owayed M. Al-Shemmari, MBBS, FACC
Alrayyan Hospital, Habib Medical Group
Riyadh, Saudi Arabia

This pre-specified subgroup analysis from the TWILIGHT trial evaluated ticagrelor monotherapy versus dual antiplatelet therapy (DAPT) with ticagrelor plus aspirin on bleeding in high-risk patients undergoing PCI after an acute coronary syndrome (ACS), unstable angina, or non-STEMI. Upon completion of a 3-month period of DAPT, event-free patients were randomized in a double-blind manner to aspirin or placebo with continuation of ticagrelor for an additional 12 months. The primary endpoint was Bleeding Academic Research Consortium (BARC) types 2, 3, or 5 bleeding, and the key secondary endpoint was the composite of all-cause death, MI, or stroke. A total of 4,614 patients was enrolled in this trial; 34.2% had 1-3 clinical and angiographic high-risk features, 48.6% had 4-5, and 17.3% had 6 or more. Mean age was 64.2 years, one-third had diabetes, and one-quarter had a previous MI. Treatment adherence was high.

The rate of BARC 2, 3, or 5 bleeding was 7.6% in the DAPT arm with ticagrelor and aspirin compared with 3.6% in the ticagrelor monotherapy arm (hazard ratio [HR] 0.47; 95% confidence interval [CI], 0.36-0.61). These results remained consistent irrespective of the number of risk factors the patients had (p for interaction = .69) and were in line with the main TWILIGHT trial. Other pre-specified bleeding endpoints favored the ticagrelor monotherapy: BARC 3 or 5 bleeding in 0.8% vs. 2.1% (p < .0001), and TIMI major bleeding in 0.5% vs. 1% (p = .08). The ischemic outcome of death, MI, or stroke was similar in the monotherapy and DAPT arms: monotherapy was 4.3%, and DAPT was 4.4% (HR = 0.97; 95% CI, 0.74-1.28):

  • Cardiovascular death, MI, or ischemic stroke: monotherapy = 4%; DAPT = 4.2% (p = .77)
  • Ischemic stroke: monotherapy = 0.5%; DAPT = 0.3% (p = .21)
  • All-cause death: monotherapy = 1%; DAPT = 1.5% (p = .14)
  • Any MI: Both groups = 3.1% (p = .99)
  • Definite or probable stent thrombosis: monotherapy = 0.4%; DAPT = 0.5% (p = .38)

Similar to TWILIGHT, several other recent trials (GLOBAL LEADERS [GLOBAL LEADERS: A Clinical Study Comparing Two Forms of Antiplatelet Therapy After Stent Implantation], SMART-CHOICE [P2Y12 Inhibitor Monotherapy vs. Dual Antiplatelet Therapy in Patients Undergoing Percutaneous Coronary Intervention], and STOPDAPT-2 [Short and Optimal Duration of Dual Antiplatelet Therapy After Everolimus-Eluting Cobalt-Chromium Stent-2]) evaluated the safety of monotherapy (not only with ticagrelor, but also with clopidogrel and prasugrel). The cumulative body of evidence suggests that discontinuation of aspirin is safe even in ACS. All the trials demonstrated lower bleeding events and similar major adverse cardiac events. A limitation of the TWILIGHT study is that it excluded STEMI. TWILIGHT, however, is the only one with a double-blind design. Although TWILIGHT included high-risk individuals and diabetics, there are many patients who fall outside the study population such as older patients, those who require triple therapy, those with complex interventions such as bifurcations and chronic total occlusions, and shock patients. Additionally, the response of potent antiplatelet agents such as ticagrelor is variable in certain ethnic groups such as Asian patients. Another group, women, have higher bleeding events after PCI on DAPT, and the current risk scores don't capture that risk. Furthermore, the discontinuation rate of ticagrelor is higher in women, as noted in the GLOBAL LEADERS sub-study. Overall, the results of this trial are reassuring and do suggest that we should be considering monotherapy in the majority of our ACS patients. Practical issues such as cost and adherence may be a limitation with some patients who must be identified.

A Propensity-Matched Comparison From the FRANCE-TAVI Registry

By Neal S. Kleiman, MD, FACC
Houston Methodist Hospital
Houston, TX

Van Belle et al. report the hospital and 2-year outcomes of patients who underwent transcatheter aortic valve replacement (TAVR) with either the Edwards balloon-expandable SAPIEN (Edwards Lifesciences; Irvine, CA) or the Medtronic self-expanding CoreValve (Medtronic, Dublin, Ireland) among patients enrolled in the FRANCE-TAVI registry. Of the 12,141 patients in the registry, propensity matching was performed on 3,910 patients receiving the SAPIEN valve and an equal number of patients treated with the CoreValve. The co-primary endpoint of paravalvular regurgitation (PVR) ≥ moderate in severity or death was higher among CoreValve-treated patients (19.8% vs. 11.9%; risk ratio [RR] 1.68; 95% CI, 1.46-1.91)]; the other co-primary endpoint, mortality at 2 years, was also higher (29.8% vs. 26.6%; HR 1.17; 95% CI, 1.06-1.29). In particular, PVR detected in hospital was more frequent among patients treated with the balloon-expandable valve (15.5% vs. 18.3%; HR 1.86; 95% CI, 1.59-2.18). Mean transvalvular gradients were lower among patients receiving the CoreValve (7 mm vs. 10 mm; p < 0.0001). When inverse probability weighting was performed, which allows inclusion of more patients, the results were similar. The study was published simultaneously in Circulation. Previous studies had reported that moderate PVR was associated with higher intermediate-term mortality, and some but not all studies had made the similar observations for mild PVR. The authors of this study inferred that PVR was the major mechanism of the observed mortality difference.

The findings are certainly intriguing because the rates of TAVR continue to increase, currently exceeding the rates of surgical aortic valve replacement, and TAVR is currently approved across the whole spectrum of surgical risk. Although the study is rich in detail and the findings internally consistent, it is important to recognize the limitations of such an approach. As the authors mention, the data are registry-derived rather than the results of a randomized investigation. Thus, the reasons for patient selection for a particular valve are not entirely known and unrecognized confounders, in particular aortic root anatomy and calcification, and operator experience with the particular valve are difficult to quantify. The applicability of this analysis to current practice may also be limited because the latter portion of the registry included the latest iteration of the SAPIEN valve (S3) but not the more recently released CoreValve Evolut Pro, which contains an external wrap designed to reduce PVR. Quantification of aortic regurgitation is site-reported rather than core-laboratory-determined and is severalfold higher than the rates reported in clinical trials. These findings may represent outcomes in the all-comers nature of registry populations compared with the rigorously selected subjects in clinical trials but may also reflect different calibration of the standards used to define the degree of regurgitation. There is also a bit of internal inconsistency that needs to be understood as far as the mechanism of differential mortality is concerned. Although no one knows for certain the time point at which PVR (other than a severe leak) would be expected to become manifest as a risk for death, one might speculate that the magnitude of its effect would increase during the duration of follow-up. However, the excess risk associated with self-expanding transcatheter heart valve implantation was present only during the first 3 months of observation. Consequently, the proportional hazard assumption was not met.

How these differences will ultimately play out over the longer term, when the competing risks of PVR and residual gradient progress particularly in lower-risk patients who are expected to have a 10-or-more-year survival horizon after aortic valve replacement, is yet to be seen. The two valves studied are by far the most commonly used for TAVR. As the authors of this study point out, the findings certainly call for a large-scale randomized trial if the question of valve selection is to be addressed completely.

AUGUSTUS (Antithrombotic Therapy After Acute Coronary Syndrome or PCI in Atrial Fibrillation)

By Matthew William Sherwood, MD, MHS, FACC
INOVA Heart and Vascular Institute
Falls Church, VA

The AUGUSTUS trial was a unique study designed to evaluate two clinical questions independently:

  1. The benefit of apixaban versus vitamin K antagonist (VKA) therapy in patients with atrial fibrillation (AF) undergoing PCI
  2. The benefit of aspirin versus placebo in the same population, with all patients on a background of P2Y12 inhibitor therapy

The results of the trial indicated that there was significant benefit with lower major and clinically relevant nonmajor bleeding rates for apixaban versus VKA (HR 0.69; 95% CI, 0.58, 0.81) in this population and significantly lower bleeding rates for placebo versus aspirin (HR 0.53; 95% CI, 0.45, 0.63). Although the trial was not powered to detect differences in ischemic event rates such as MI or stroke, there was a small but significant reduction in stroke (HR 0.50; 95% CI, 0.26, 0.97) and rehospitalization rates (HR 0.83; 95% CI, 0.74, 0.93) with apixaban versus warfarin but no difference in outcomes with aspirin versus placebo. The AUGUSTUS trial was performed in the context of several trials (PIONEER AF-PCI [An Open-Label, Randomized, Controlled, Multicenter Study Exploring Two Treatment Strategies of Rivaroxaban and a Dose-Adjusted Oral Vitamin K Antagonist Treatment Strategy in Subjects With Atrial Fibrillation Who Undergo Percutaneous Coronary Intervention], RE-DUAL PCI [Randomized Evaluation of Dual Antithrombotic Therapy With Dabigatran vs. Triple Therapy With Warfarin in Patients With Nonvalvular Atrial Fibrillation Undergoing Percutaneous Coronary Intervention], and ENTRUST-AF PCI [Edoxaban-Based Antithrombotic Regimen in Patients With Atrial Fibrillation]) showing that direct oral anticoagulants with P2Y12 inhibitors are safer than traditional triple therapy (VKA + P2Y12 inhibitor + aspirin) for patients with AF undergoing PCI. The results of this trial further validated the premise that triple therapy is associated with significant excess bleeding and that the standard of care should be direct oral anticoagulants with a P2Y12 inhibitor alone. Meta-analyses that include data from the AUGUSTUS, PIONEER-AF PCI, and RE-DUAL PCI trials have shown that the effect is consistent across studies and have provided pooled evidence that there is no increase in ischemic events with a dual-therapy versus triple-therapy regimen. Although the guidelines have not yet changed, both European and American position statements have gathered consensus and recommended the use of dual therapy with a direct oral anticoagulant and clopidogrel as the standard approach for patients with AF undergoing PCI. The optimal duration of therapy remains an outstanding question, but it seems that 12 months of combined therapy is sufficient, after which anticoagulation alone would be the preferred approach.

Mortality and Bleeding Among Patients With Acute Myocardial Infarction Complicated by Cardiogenic Shock Undergoing Percutaneous Coronary Intervention With Impella® vs Intra-Aortic Balloon Pump

By Alexander George Truesdell, MD, FACC
Virginia Heart/INOVA Heart and Vascular Institute
Falls Church, VA

Following the adoption of early revascularization for acute MI, mortality rates for acute MI with cardiogenic shock (CS) decreased from near 90% to less than 50%. In decades since, in-hospital survival rates have plateaued while the incidence of acute MI with CS has increased, despite improvements in pharmacotherapy, PCI techniques and technology, and door-to-balloon times. More recently, cardiac surgeons, cardiac intensivists, HF specialists, and interventional cardiologists have increasingly looked to mechanical circulatory support (MCS) devices to improve outcomes in acute MI and acute decompensated HF with CS and to facilitate ever more complex and higher-risk (HR) PCI. Unfortunately, clinical trials evaluating drugs, devices, and best practices for acute MI with CS have been both challenging to conduct and slow to enroll, with fewer than 2,500 patients studied in prospective randomized controlled trials (RCTs) since the landmark 1999 SHOCK (SHould we emergently revascularize Occluded Coronaries for cardiogenic shocK) study. Two observational studies presented at the recent American Heart Association 2019 Scientific Sessions by Dhruva et al. and Amin et al. offer a limited snapshot of contemporary IABP and Impella (Abiomed; Danvers, MA) use and appropriately raise important safety and efficacy questions but unfortunately do not provide any reliable answers.

Dhruva et al. presented "Mortality and Bleeding Among Patients With Acute Myocardial Infarction Complicated by Cardiogenic Shock Undergoing Percutaneous Coronary Intervention With Impella® vs Intra-Aortic Balloon Pump" comparing MCS utilization among a group of 28,304 patients in the NCDR Cath-PCI and Chest Pain-MI Registries undergoing PCI for acute MI with CS between 2015 and 2017. The authors reported an increase in Impella use for PCI with acute MI with CS from 3.5% to 8.7% and a parallel decrease in IABP use from 32.1% to 27.3%, which remains high despite the neutral results of the randomized 2012 IABP-SHOCK II (Intraaortic Balloon Pump in Cardiogenic Shock II) trial. The authors concluded that, among a small cohort of 1,680 propensity-matched patients with acute MI with CS from the larger group, use of Impella was associated with higher rates of death and bleeding complications compared with IABP. Accordingly, Dhruva et al. call for additional evidence to define optimal management of patients with acute MI with CS and the role of Impella in clinical practice.

Amin et al. presented "The Comparative Effectiveness and Costs of Impella vs. Intra-Aortic Balloon Pump in the United States" analyzing 48,306 patients from the Premier Healthcare Database undergoing PCI with MCS between 2004 and 2016. The authors examined trends and variations in rates of Impella use (for both CS and other indications) as well as associated clinical outcomes including in-hospital mortality, bleeding requiring transfusion, acute kidney injury, stroke, length of stay, and hospital costs. They noted an increase in Impella use over time, reaching 31.9% of total MCS use in 2016. The authors also observed a very wide variation in both Impella use and clinical outcomes across 432 hospitals. They further noted that Impella use was associated with higher rates of death, bleeding, and stroke. Amin et al. concluded that more data are needed to define the appropriate role of MCS in patients undergoing PCI and that the significant variability in Impella use and its associated outcomes underscores the need to better define the appropriate use of MCS devices with adequately powered RCTs and prospective real-world evidence.

Both studies describe a young patient population (average age 65 years old) with high mortality rates, highlighting our collective interest in improving outcomes in the growing acute MI with CS population. Amin et al. did not differentiate between CS and non-CS indications for Impella use in their report and examined groups that were not entirely well-matched, with the Impella group notable for higher rates of HF, diabetes, chronic kidney disease, multivessel disease, bifurcation lesions, chronic total occlusions, and atherectomy. Neither study examined the majority of PCI and CS populations treated without MCS nor patients treated with alternate forms of MCS such as extracorporeal membrane oxygenation.

Looking forward, there is great need for high-quality evidence for MCS device use in acute MI with CS, acute decompensated HF with CS, and higher-risk PCI, both from RCTs and all-comer registries (and perhaps registry-based RCTs). There is likewise need for better phenotyping of diverse shock etiologies and stages to effect more appropriate patient selection and improve "apples-to-apples" comparisons, particularly in such a complex and heterogenous patient population with many paths to morbidity and mortality. We ought to also examine the utility of detailed hemodynamic assessments and specific treatment pathways and protocols for all pharmacologic and device-based therapies as well as spoke-and-hub systems of care. Finally, there should be more emphasis on a spectrum of outcomes beyond in-hospital mortality, to include longer-term mortality, infarct size, HF readmissions, and other important patient-centered outcomes. Until then, most contemporary "Level 1" cardiac care centers will (and perhaps should) continue to employ a wide array of plug-and-play technologies like IABP, Impella, TandemHeart (CardiacAssist, Inc.; Pittsburgh, PA), extracorporeal membrane oxygenation, and others, guided by home-grown triage, treatment, and escalation protocols employed by increasingly experienced multidisciplinary shock teams.

COMPLETE (Complete Versus Culprit-Only Revascularization Strategies to Treat Multivessel Disease After Early PCI for STEMI) OCT Substudy

By Michael N. Young, MD, RPVI
Dartmouth-Hitchcock Medical Center
Geisel School of Medicine at Dartmouth
Lebanon, NH

In October 2019, Dr. Mehta and colleagues published findings of the COMPLETE trial in The New England Journal of Medicine. The COMPLETE trial, a large, international, multicenter clinical trial, randomized over 4,000 patients with STEMI with prior successful culprit-lesion PCI to complete revascularization versus guideline-directed medical therapy. At a median follow-up of 3 years, complete revascularization was superior compared with medical therapy for a first co-primary outcome of composite cardiovascular death or MI (HR 0.74; 95% CI, 0.60-0.91; p = 0.004), as well as the secondary co-primary endpoint that also included ischemia-driven revascularization (HR 0.51; 95% CI, 0.43-0.61; p < 0.001).

The COMPLETE trial investigators conducted a nested intracoronary imaging substudy using optical coherence tomography (OCT). Pinilla-Echeverri and colleagues presented their OCT results of non-culprit coronary lesions (NCLs) of patients within the COMPLETE trial. Specifically, they hypothesized that thin-cap fibroatheroma—a known feature of vulnerable plaque morphology—may be more prevalent in obstructive versus non-obstructive NCLs. Trial patients randomized to complete revascularization or PCI intended for a non-culprit lesion underwent multivessel OCT imaging (n = 93) that encompassed both obstructive and non-obstructive NCLs. The investigators discovered that obstructive versus non-obstructive NCLs were more likely to be vulnerable based on a higher prevalence of thin-cap fibroatheroma (58/150 vs. 74/275; 35.4% vs. 23.2%; p = 0.022). Furthermore, obstructive NCLs with thin-cap fibroatheroma possessed more features of plaque vulnerability than NCLs without thin-cap fibroatheroma (e.g., number of lipid quadrants, lipid arc, macrophages, and cholesterol crystals). Perhaps most compelling is the fact that 47.3% of patients studied possessed an obstructive NCL that also exhibited vulnerable plaque.

The OCT substudy of the COMPLETE trial provides valuable insight into the potential benefits of complete revascularization. Although this intracoronary imaging sub-investigation was not necessarily powered to explain the overarching clinical endpoints of the COMPLETE trial, it offers a potential patho-anatomical explanation as to why those patients with STEMI who were randomized to complete revascularization (i.e., PCI of non-obstructive NCLs) fared better than those who were treated medically. Furthermore, these findings add to the growing body of literature that supports the utility of intracoronary imaging for understanding coronary disease pathology, particularly in the context of percutaneous revascularization.

Keywords: AHA Annual Scientific Sessions, AHA19, Acute Coronary Syndrome, Anticoagulants, Arrhythmias, Cardiac, Aspirin, Atrial Fibrillation, Fibrinolytic Agents, Hemorrhage, Myocardial Revascularization, Percutaneous Coronary Intervention, Secondary Prevention, Thrombosis, Vitamin K, Warfarin, Transcatheter Cardiovascular Therapeutics, Acute Coronary Syndrome, Cardiac Surgical Procedures, Constriction, Pathologic, Coronary Artery Bypass, Coronary Artery Disease, Fractional Flow Reserve, Myocardial, Hemorrhage, Hospitalization, Myocardial Infarction, Myocardial Revascularization, Percutaneous Coronary Intervention, Thrombosis, Acute Coronary Syndrome, Adenosine, Angiography, Aspirin, Atherectomy, Brain Ischemia, Coronary Artery Disease, Diabetes Mellitus, Drug-Eluting Stents, Hemorrhage, Myocardial Infarction, Myocardial Ischemia, Percutaneous Coronary Intervention, Platelet Aggregation Inhibitors, Renal Insufficiency, Stroke, Thrombosis, Troponin, Vascular Diseases, Acute Coronary Syndrome, Angina, Stable, Angina, Unstable, Cardiac Catheterization, Constriction, Pathologic, Coronary Angiography, Coronary Artery Bypass, Coronary Artery Disease, Exercise Test, Heart Arrest, Heart Failure, Magnetic Resonance Imaging, Myocardial Infarction, Myocardial Ischemia, Percutaneous Coronary Intervention, Renal Insufficiency, Chronic, Secondary Prevention, Stroke Volume, Tomography, Acute Coronary Syndrome, Angina Pectoris, Angina, Unstable, Cardiac Catheterization, Coronary Angiography, Coronary Artery Bypass, Coronary Artery Disease, Exercise Test, Heart Arrest, Heart Failure, Kidney Diseases, Magnetic Resonance Imaging, Myocardial Infarction, Myocardial Ischemia, Myocardial Revascularization, Percutaneous Coronary Intervention, Renal Dialysis, Renal Insufficiency, Chronic, Secondary Prevention, Stroke, Stroke Volume


< Back to Listings