The Take Home: AHA Scientific Sessions
The American Heart Association Scientific Sessions had several presentations with major implications for cardiology, most notably the ISCHEMIA and ISCHEMIA-CKD trials, considered to be the most important findings for management of stable ischemic heart disease in more than a decade.
Healio and Cardiology Today were onsite and gathered numerous perspectives from experts, including Kirk N. Garratt, MD, MSc, MSCAI, from Christiana Care and past president of the Society for Cardiovascular Angiography and Interventions; Ajay J. Kirtane, MD, SM, from NewYork-Presbyterian Hospital and Columbia University Irving Medical Center; Mikhail Kosiborod, MD, FACC, FAHA, from Saint Luke’s Mid America Heart Institute and University of Missouri-Kansas City School of Medicine; Cardiology Today Next Gen Innovator Dharam J. Kumbhani, MD, SM, from University of Texas Southwestern Medical Center; Cardiology Today Editorial Board Member Erin D. Michos, MD, MHS, FAHA, from Johns Hopkins Ciccarone Center for the Prevention of Heart Disease; Athena Poppas, MD, FACC, FASE, from Rhode Island, The Miriam and Newport Hospitals and vice president of the American College of Cardiology; Paul M. Ridker, MD, MPH, FACC, FAHA, from Brigham and Women’s Hospital and Harvard Medical School; Cardiology Today Next Gen Innovator Binita Shah, MD, MS, from NYU School of Medicine and VA New York Harbor Healthcare System; and B. Hadley Wilson, MD, FACC, from Atrium Health’s Sanger Heart & Vascular Institute.
Read below for more.
ISCHEMIA and ISCHEMIA-CKD
Kirtane: The big question in the main ISCHEMIA trial was whether there could be a benefit or a harm to an invasive strategy in patients with stable ischemic heart disease.
In 5,179 patients, the primary outcome of CV death, MI, hospitalization for unstable angina, HF or resuscitated cardiac arrest at 4 years occurred in 13.3% of the invasive group vs. 15.5% of the conservative group (absolute difference, –2.2%; 95% CI, –4.4 to 0; adjusted HR = 0.93; 95% CI, 0.8-1.08), but the curves crossed at approximately 2 years, favoring the conservative group early and the invasive group late. The outcome of benefit of invasive therapy on Seattle Angina Questionnaire summary score favored the invasive group in the overall population at 3 months (posterior mean = 4.1; 95% CI, 3.2-5), 1 year (posterior mean = 4.2; 95% CI, 3.3-5.1) and 3 years (posterior mean = 2.9; 95% CI, 2.2-3.7).
In ISCHEMIA-CKD, which included 777 patients with chronic kidney disease who were ineligible for the main trial, the results for CV events were similar to that of the main trial, but quality of life data did not favor either group.
The answer is nuanced because there is an early procedural harm. It’s probably from the revascularization. But that is offset by a small late benefit in terms of a reduction in MI. Over 4 years, there was a reduction in CV death and MI with the invasive approach. That suggests there are patients for whom the invasive approach can prevent heart events. The trial was not fully powered for to look at types of heart events, so we have to be circumspect about this. But there are other streams of data, from meta-analyses and FAME 2, showing a benefit in spontaneous MI. This is an important finding. But we have to be cognizant of the procedural risks for every patient.
Determining who would benefit from the invasive strategy includes taking account of how symptomatic the patients are at baseline. In clinical practice, if the patient does not have a lot of symptoms, I do CT angiography to rule out bad disease, and try medical therapy. But if the patient has a lot of symptoms, the invasive approach may be best.
Garratt: The main ISCHEMIA trial showed that, consistent with earlier studies, use of PCI early for management of patients with moderate to severe ischemia is not required to provide benefit as measured by the endpoints of the trial. Good medical therapy is what ought to be used first as a means of managing patients with this heart problem.
We did see, though, that use of PCI early was safe, with no indication of hazard associated with a strategy of early PCI use. Interestingly, there was a signal of lower rates of spontaneous MI within 4 years after treatment for patients who got PCI rather than medical therapy alone. That’s consistent with what we saw in the FAME 2 trial and wasn’t too much of a surprise, but was a gratifying finding from the perspective of interventional cardiology. Having that information now will let us have more meaningful conversations with our patients about how PCI might be able to benefit them.
The big news for interventionalists was the quality of life findings. Those findings were key because in American practice today we don’t typically offer PCI to patients with the promise of lower rates of death or even lower rates of MI. We are offering this therapy to patients primarily as a way of improving their quality of life. The data show that PCI is superior to medical therapy alone for anginal relief across anginal symptom classes. That’s also going to let us have better conversations with patients about just how PCI might be able to help them.
For ISCHEMIA-CKD, the similar message is that there is not an indication that PCI is needed in the early management of patients with CKD and moderate to severe ischemia. Rather, medical therapy alone is sufficient to keep them safe. It is much smaller than the main study, so the reliability of the findings will be less. The take-home message is we need to be prudent in the use of PCI in these very high-risk patients. These patients were at high risk for subsequent CVD events, so there was some hope of benefit from aggressive treatment. That was not forthcoming. These patients are also at high risk for procedural complications. While most patients from the invasive group did not have worsening kidney function, that is always a point of anxiety for interventional cardiologists. Most interventional cardiologists, if they’re not already conservative about offering angioplasty in this population, will be going forward.
Kosiborod: A key secondary objective of DAPA-HF, in which dapagliflozin (Farxiga, AstraZeneca) lowered risk for CV death and HF hospitalization compared with placebo in patients with HF with or without diabetes, was to examine the effects of dapagliflozin 10 mg on the 23-item Kansas City Cardiomyopathy Questionnaire (KCCQ) overall summary score.
In 4,443 patients, consistent effects of dapagliflozin compared with placebo across KCCQ total symptom score tertiles for reducing the primary endpoint of CV death or worsening HF, with HRs of 0.7 (95% CI, 0.57-0.86) in the lowest tertile, 0.77 (95% CI, 0.61-0.98) in the middle tertile and 0.62 (95% CI, 0.46-0.83) in the highest tertile. At 8 months, participants treated with dapagliflozin experienced a greater improvement in mean KCCQ total symptom score (mean, 2.8 points), clinical summary score (mean, 2.5 points) and overall summary score (mean, 2.3 points; P < .0001 for all) compared with those assigned placebo. In addition, the treatment effect of dapagliflozin was consistent across age groups.
Dapagliflozin not only reduced CV death and hospitalizations, but also significantly improved symptoms, physical limitations and quality of life in patients with heart failure with reduced ejection fraction. These effects were substantial and clinically important.
In our view, dapagliflozin offers a new approach, not just for reducing death from hospitalizations, but improving symptoms, physical limitations and quality of life in patients with HF with reduced ejection fraction.
Ridker: The COLCOT trial is an external and independent confirmation that the inflammation hypothesis is real. That is really crucial.
In 4,745 adults recruited within 30 days after MI, the primary endpoint — CV death, resuscitated cardiac arrest, MI, stroke or hospitalization for angina leading to coronary revascularization — occurred in 5.5% of participants in the colchicine arm and 7.1% of participants in the placebo arm, for an HR of 0.77 (95% CI, 0.61-0.96). Colchicine also conferred risk reduction in CV death, resuscitated cardiac arrest, MI, stroke and urgent hospitalization for angina.
When we presented our CANTOS data 2 years ago, it was the first study to support the inflammation hypothesis. Like any field, we need replication, and this clearly provides it. It is going to dictate what is going to happen in atherosclerosis. Down the road, I would like to see cardiologists treating all atherosclerosis patients with some combination of powerful lipid-lowering drugs and powerful anti-inflammatory drugs. It is clear now that we need to do both. One can imagine using colchicine in high-risk patients in whom we are already doing the recommended things. Right now, that is probably the group I would consider using this in because the evidence is positive, and I don’t have an alternative anti-inflammatory drug that is proven. Additionally, there will be at least three more large colchicine trials coming out in the next year or 2. If they are all consistent, this could well become a new therapy for atherosclerosis, so it is very exciting.
Shah: COLCHICINE-PCI showed for the first time that colchicine can prevent a rise in blood markers of vascular inflammation during an acute injury.
The study, which I presented, evaluated oral administration of 1.8 mg of colchicine or placebo to examine the effects on myocardial injury and biomarker evidence of PCI-related inflammation in 714 patients with suspected ischemic heart disease or ACS referred for coronary angiography with possible PCI.
The primary outcome of PCI-related myocardial injury did not differ between patients administered colchicine or placebo, at 57.3% in the colchicine group compared with 64.2% in the placebo group (P = .19). However, there was significant attenuation in the rise of interleukin-6 (P = .02) and high-sensitivity C-reactive protein concentration (P = .001) at 22 to 24 hours after PCI in those who received colchicine. There was no difference in interleukin-1.
More work is needed to determine the optimal dosing and timing regimen of colchicine administration in patients with CAD and patients undergoing PCI. We saw inflammatory markers decrease around the 24-hour time point post-PCI, so an earlier start to preprocedural colchicine regiment warrants further investigation.
GALILEO and GALILEO-4D
Wilson: We do not know the optimal anticoagulant or antiplatelet regimen for patients who have had transcatheter aortic valve replacement. Aspirin and clopidogrel have been the most common agents prescribed post-TAVR for at least 3 months. However, there have been concerns that a post-TAVR anticoagulation regimen might be more effective.
The hypothesis in the GALILEO trial of 1,644 patients who underwent TAVR was that there would be fewer deaths and thromboembolic events in the group assigned rivaroxaban (Xarelto, Janssen/Bayer). But actually, there were more deaths and thromboembolic complications in the rivaroxaban group compared with the standard aspirin and clopidogrel regimen (9.8 per 100 person-years vs. 7.2 per 100 person-years; HR = 1.35; 95% CI, 1.01-1.81).
This study shows the regimen of rivaroxaban plus aspirin after TAVR is not beneficial and is actually quite harmful for patients. Questions must be asked about whether this was the right dose and the right anticoagulant. Further studies could help determine if a lower dose of rivaroxaban may be safer and provide equal or better outcomes than standard care for patients who had TAVR. Right now, the results endorse aspirin plus clopidogrel for at least 3 months post-TAVR.
GALILEO-4D, a substudy of 231 patients who had TAVR and underwent four-dimensional CT, pushes on the theory of using direct oral anticoagulants in post-TAVR patients. We have observed other recent studies showing there is leaflet thrombus accumulation and leaflet thickening post-TAVR, particularly early on, despite the standard treatment of aspirin plus clopidogrel. Indeed, the study showed there was less reduced leaflet motion (2.1% of the rivaroxaban group compared with 10.9% of the antiplatelet group; difference, –8.8 percentage points; 95% CI, –16.5 to –1.9) and reduced thickening (12.4% of the rivaroxaban group had thickening of at least one leaflet compared with 32.4% of the antiplatelet group; difference, –20 percentage points; 95% CI, –30.9 to –8.5), especially early on, with an oral anticoagulant, but the clinical results from the larger GALILEO study did not support that made a clinical difference in avoidance of more serious complications. And in GALILEO-4D, the leaflet motion and thickening results evened out over time.
These two studies point in different directions. GALILEO-4D suggests oral anticoagulation will be beneficial to patients post-TAVR, but the larger GALILEO study showed the opposite. It does not necessarily mean we should discard all anticoagulants and all anticoagulant regimens for this population. But right now, we should stick with the standard of care with aspirin and clopidogrel.
Michos: We are in a new era of lipid therapy. We used to have nothing available but statins and ezetimibe. Now, there are a number of new agents that have emerged or are on the near horizon, like inclisiran (The Medicines Company). The twice-a-year dosing is an exciting option to give patients, as compliance with medical therapy is difficult for many. Even with statins, which are inexpensive and effective, studies have shown that there is variable compliance. And, patients who are nonadherent have worse outcomes. I liken this to the flu shot; many people get their flu shot once per year. With this therapy, it's like getting your flu shot twice per year.
In the trial of 1,561 patients with ASCVD and elevated LDL despite maximum tolerated doses of LDL-lowering therapies, inclisiran decreased LDL by 58% at 17 months (P < .0001), which is remarkable.
Inclisiran also appears to be safe, based on the data that have been reported thus far. Injectable therapies, as we have seen with the PCSK9 inhibitors, can have some injection-site reactions. In ORION-10, the injection-site reactions were mostly mild and transient. Price is important when it comes to new therapy, so we wait to see what the price of this therapy at this dosing would be. The big thing I'm holding out for is the ORION-4 CV outcomes trial, which will provide information on whether inclisiran reduces CV events. I am a strong believer in the LDL hypothesis, and I think if we can lower LDL, we are going to lower events.
Kumbhani: Among 5,763 patients with ACS from the TWILIGHT trial who underwent PCI, ticagrelor monotherapy reduced bleeding risk compared with dual antiplatelet therapy, with no additional ischemic risk.
At 1 year after randomization, which occurred 3 months after PCI, the rate of BARC 2, 3 or 5 bleeding was 7.6% in those assigned DAPT with ticagrelor (Brilinta, AstraZeneca) and aspirin compared with 3.6% in those assigned ticagrelor monotherapy (HR = 0.47; 95% CI, 0.36-0.61). The primary ischemic outcome of death, MI or stroke at 1 year was almost identical in both groups (monotherapy, 4.3%; DAPT, 4.4%; HR = 0.97; 95% CI, 0.74-1.28).
This is a very important subgroup analyses from this landmark trial because there were earlier data showing if you use shorter duration of DAPT in patients with ACS, you may pay a penalty. One, SMART-DATE, showed a concerning MI signal when DAPT was stopped at 6 months compared with 12 months. Against that background, TWILIGHT-ACS is reassuring that the bleeding benefit seen in the main trial is also present in the ACS subset, which is about two-thirds of the entire population. It was not powered for ischemic endpoints, but at least we didn’t see any obvious penalty in hard endpoints or stent thrombosis.
One distinction between SMART-DATE and TWILIGHT is that patients were kept on ticagrelor monotherapy in TWILIGHT till 12 months; it was altogether stopped at 6 months in SMART-DATE and aspirin monotherapy was continued. Another important nuance is that patients with STEMI were not included in TWILIGHT. One can imagine that those patients would have the highest thrombotic risk. These results are reassuring in non-STEMI, but patients with STEMI need to be further studied.
Poppas: We need to be able to compare different devices, and head-to-head comparisons can be challenging. That being said, there may be reasons for operator choices and there may be patient variables that may make somebody choose one device over another.
In this retrospective registry-based study, the researchers matched 3,910 patients who received a balloon-expandable valve (Sapien family of products, Edwards Lifesciences) with 3,910 patients who received a self-expanding valve (CoreValve family of products, Medtronic). The first coprimary outcome of moderate or worse paravalvular regurgitation and/or in-hospital mortality was higher in the self-expanding group compared with the balloon-expandable group (19.8% vs. 11.9%; RR = 1.68; 95% CI, 1.46-1.91). The self-expanding group had higher incidence of both moderate or worse paravalvular regurgitation (15.5% vs. 8.3%; RR = 1.9; 95% CI, 1.63-2.22) and in-hospital mortality (5.6% vs. 4.2%; RR = 1.34; 95% CI, 1.07-1.66). The second coprimary outcome of 2-year mortality was also higher in the self-expanding group (29.8% vs. 26.6%; RR = 1.17; 95% CI, 1.06-1.29).
The authors should be commended for doing a great job with more than 25 variables to perform propensity matching. We know that there still can be confounding and bias and that there is no perfect matching that can be attained. But using a contemporaneous data set and knowing that each of the devices was improved on after that time point helps inform us a little better.
If we think about paravalvular regurgitation being important for morbidity and mortality, the balloon-expandable valves seemed to perform better in each generation. Further research should be done as to why, and what the relationship is between paravalvular regurgitation and mortality. Is paravalvular regurgitation a cause of mortality or a marker? This retrospective study cannot answer that, but provokes us to look at it more closely. Another question that remains is if there is a difference between these valves in durability.
Kumbhani: In an OCT substudy of the COMPLETE trial, we learned that that 47% of patients had obstructive nonculprit lesions with vulnerable plaque, which could explain why complete revascularization conferred better outcomes than culprit lesion-only revascularization in the main trial.
In the cohort of 93 patients, the prevalence of thin-cap fibro atheroma per lesion was 35.4% in obstructive lesions and 23.2% in nonobstructive lesions (P = .022). In this population, 47.3% had an obstructive nonculprit lesion with vulnerable plaque, 20.4% had a nonobstructive nonculprit lesion with vulnerable plaque and 32.3% had no nonculprit lesions with vulnerable plaque.
This was an interesting analysis in a small population. It mechanistically confirms what we now seem to think about why the biology of why multivessel PCI may be beneficial in patients with STEMI who are not in cardiogenic shock. There are likely more unstable lesions than can be detected angiographically. It was helpful to corroborate that, even if it was in a small subset. – Reporting by Erik Swain, Darlene Dobkowski, Katie Kalvaitis and Regina Schaffer
Disclosures: Garratt, Kumbhani, Michos, Poppas and Wilson report no relevant financial disclosures. Kirtane reports he received institutional research grants from Abbott Vascular, Abiomed, Boston Scientific, Cardiovascular Systems Inc., Medtronic, Philips and ReCor Medical. Kosiborod reports he has received grants, honoraria and other research support from Amarin, Amgen, Applied Therapeutics, AstraZeneca, Bayer, Boehringer Ingelheim, Eisai, Eli Lilly, GlaxoSmithKline, Glytec, Janssen, Merck, Novartis, Novo Nordisk and Sanofi. Ridker reports he received research grants from Kowa and Novartis, is listed as a co-inventor on patents related to use of inflammatory biomarkers in CVD licensed to AstraZeneca and Siemens, and serves as a consultant/advisory board member for Janssen and Novartis. Shah reports she received grant funding from the VA Office of Research and Development and the NIH/NHLBI to study colchicine in acute MI and PCI.