Mandrola’s Top 10 Cardiology Stories of 2021 – Medscape

1. Good and Bad News in Obesity

A sobering research letter in November reported that 1 in 3 young adults in the United States are obese. The pandemic intensified the importance of combating obesity, as the risk for severe COVID-19 has a linear relationship with body mass index.

The good news is that numerous randomized controlled trials published this year found that the glucagon-like peptide-1 (GLP-1) analogue semaglutide induced significant and sustained weight loss. In STEP I, a randomized controlled trial of patients with obesity without diabetes, semaglutide resulted in a 15-kg weight loss from baseline. The best part of this story is that other drugs are being developed to treat patients with obesity. Both tirzepatide (a glucose-dependent insulinotropic polypeptide and GLP-1 analogue) and cagrilintide (a long-acting amylin analogue) looked promising in early-phase studies.

Given the increasing prevalence of obesity, that a weekly injection can achieve and maintain weight loss similar to that occurring with major abdominal surgery is a serious advance.

The SGLT2 inhibitor class crushed it in 2020. Dapagliflozin proved effective in reducing renal outcomesin patients with chronic kidney disease (CKD); empagliflozin joined dapagliflozin in proving to be effective in patients with heart failure with a reduced ejection fraction (HFrEF), and the combined SGLT21/2 inhibitor sotagliflozin looked impressive in patients with diabetes, heart failure, and CKD.

But the results from 2021 were less impressive. The EMPEROR-Preserved trial of empagliflozin vs placebo in patients with heart failure with a preserved ejection fraction (HFpEF) had a "positive" topline result, but it was hardly a win. The 21% reduction in the composite primary outcome of hospitalizations for heart failure and cardiovascular (CV) death was driven by fewer heart failure hospitalizations; CV death did not differ significantly. And heart failure hospitalizations accounted for less than 20% of total hospitalizations.

The problem, I think, is with the HFpEF classification. Unlike HFrEF, where the one core issue is reduced systolic function, HFpEF is essentially a disease of aging, deconditioning, and metabolism. Patients with HFpEF often have multiple competing causes of morbidity and mortality. The reason drugs have not had major effects on HFpEF is because these patients have numerous problems.

The LAAOS III trial studied the stroke preventive effects of add-on surgical closure of the left atrial appendage (LAA) in patients with a history of atrial fibrillation (AF) who were having cardiac surgery. More than 2000 patients were randomly assigned to undergo add-on surgical closure or just have their assigned cardiac surgery. Both groups received standard careincluding oral anticoagulation. Surgical LAA closure resulted in a statistically significant 2.2% absolute risk reduction in stroke or systemic embolism. There was no increase in bleeding and only 6 minutes of extra pump time.

LAAOS III gets my vote for the most important trial of the year. The benefits were clinically significant and statistically robust. When our patients with AF go for heart surgery, the surgeon should close their appendage according to the techniques used in LAAOS IIIwhich mostly entailed excision and primary closure.

Percutaneous closure of the appendage with foreign bodies placed in the left atrium continued to look dubious in 2021. A report chronicled risk factors for device-related thrombusan Achilles heel for the procedure. Two observational studies found significantly higher rates of in-hospital complications in women and in those with CKD.

Long-term results of PRAGUE-17 do not persuade me. Although the topline result of the LAA closure (Watchman [Boston Scientific] or Amulet [Amplatzer]) vs direct-acting oral anticoagulant trial prompted investigators to conclude continued noninferiority, there remained no reduction in stroke events. The slight advantage in lower bleeding rates in the device arm was countered by major procedural complicationsincluding two deaths.

Recall that appendage closure is a preventive procedure that delivers only probabilistic benefits in the future. The difference between concomitant surgical and percutaneous closure is that LAAOS III reported no upfront adverse risks, whereas the real-world data on percutaneous closure clearly show that patients who have this procedure incur serious upfront procedural risks.

In February, sacubitril/valsartan earned FDA approval for the treatment of patients with HFpEF. This was based on the PARAGON-HF trial, which compared sacubitril/valsartan to valsartan alone in patients with HFpEF. The 13% reduction in the composite endpoint of heart failure hospitalization and CV death with the ARNI drug did not reach statistical significance. After much debate at the FDA about subgroups, the approval came with vague wording: "Benefits are most clearly evident in patients with left ventricular (LV) ejection fraction (LVEF) below normal but LVEF is a variable measure, so use clinical judgment in deciding whom to treat."

This year, two trials testing sacubitril/valsartan also returned nonsignificant results. In the PARADISE-MI trial of more than 5600 patients with LV dysfunction and symptoms of heart failure after myocardial infarction (MI), sacubitril/valsartan did not significantly reduce a composite endpoint of CV death or heart failure compared with ramipril. And in the LIFE trial of 335 patients with advanced heart failure, sacubitril/valsartan (vs valsartan alone) did not significantly reduce the primary endpoint of N-terminal probrain natriuretic peptide ratio nor the secondary endpoint of days alive out of hospital and free of heart failure events.

Recall that the seminal PARADIGM-HF trial compared the ARNI drug to moderate doses of enalapril. The slew of nonsignificant trials since PARADIGM makes me wonder what would happen if that trial were redone with equivalent doses of valsartan alone.

Before 2021, trial evidence had definitively shown that adding revascularization to optimal medical therapy (or pursuing an early interventional approach) does not reduce the rate of MI or death in patients with stable coronary artery disease (CAD). Yet I am not sure this evidence has changed the mindset of the average clinician or patient. Pathos trumps logos when it comes to angina and focal lesions.

An observational study called CLARIFY could help settle our collective fears. This was a multinational longitudinal registry of more than 30,000 patients with stable CAD. More than 7000 patients had angina at baseline. Get this: in nearly 40% of patients with baseline angina, the angina disappeared without revascularization in the first year, and by 5 years, angina resolved in the majority. Only 5% of those with angina resolution had revascularization.

Senior author of CLARIFY, professor Gabriel Steg, said it well: "It's perfectly reasonable to have a trial of watchful waiting with medical treatment in patients who have angina and stable CAD because the risk of events is low, the chance this will regress is very high, and, if it regresses, they have an excellent outcome."

The idea behind AF screening is that AF is often asymptomatic; it increases the odds of stroke, and anticoagulants can reduce stroke risk. Yet the path from detecting asymptomatic AF to better clinical outcomes is full of obstaclesmisdiagnosis, overdiagnosis, downstream cascades, and bleeding from oral anticoagulants, for instance. Two randomized trials published this year suggested that AF screening isn't going to deliver big reductions in clinical outcomes.

STROKESTOP randomly assigned all 75- to 76-year-old adults in two counties in Sweden to be invited for AF screening or to usual care. The screening intervention was twice-daily single-lead electrocardiography. Over nearly 7 years of follow-up, the screened group sustained a statistically significant reduction in the primary outcome of ischemic or hemorrhagic stroke, systemic embolism, bleeding requiring hospitalization, and all-cause death events. The problem: the effect size was small at 0.23% absolute risk reduction, and the results were fragilewith a P value of .045.

The LOOP study randomly assigned older adults to receive an implantable loop recorder or usual care. The always-on recorder detected substantially more AF and led to nearly threefold higher rates of anticoagulant use. Yet, over 5 years, screening did not lead to a statistically significant reduction in the primary outcome of stroke or systemic embolism.

One caveat: the expansion of consumer-oriented rhythm-monitoring devices means that most AF screening in the future will occur whether or not a clinician orders it. These devices will turn people into patients. This will be good for the healthcare system, but the data thus far suggest little net benefit for the person-turned-patient.

Fractional flow reserve as a way to improve outcomes of percutaneous coronary intervention (PCI) failed twice in 2021. The FLOWER-MI trial randomly assigned patients with ST-segment elevation MI and multivessel disease who had undergone successful PCI of the culprit lesion to receive complete revascularization guided by FFR or angiography. The FFR group had a 32% higher rate of the composite primary outcome of all-cause death, MI, and hospitalization for urgent revascularization.

The FAME 3 trial randomly assigned patients with three-vessel CAD to undergo coronary artery bypass grafting (CABG) or FFR-guided PCI. Previous trials showing superiority of CABG vs PCI in three-vessel disease did not use FFR, so the idea was that FFR may put PCI on par with surgery. It was not to be. Rates of major adverse cardiac or cerebrovascular events at 1 year were significantly higher in the PCI arm (17.8% vs. 12.4% for CABG; P = .002).

The failure of FAME-3 is surprising only if you hold fast to a clogged-pipe frame of coronary disease. A neutral Martian who knew only the lack of evidence that revascularization improves outcomes over medical therapy in stable patients would ask why anyone thought defining the functional significance of a focal blockage would be important.

A medical trial typically produces average event rates in a treatment arm vs a control arm. We do the experiment only once, so we have to look at the event rates and decide whether the results were due to the play of chance or to true differences.

The convention now is to pretend there is no prior information and no differences in the two arms (the null hypothesis), and then measure how strange the data are given these assumptions (the P value). But that's not what clinicians want to know. We want to know the probability that the active arm is beneficial given the prior information and the dataas if it were a stress test.

This year, the TIMI research group from Harvard analyzed an important study using both frequentist and Bayesian methods. This was a meta-analysis of the trials of PCI vs CABG for left main CAD.

They first reported mortality rates in the typical way: the rate of death was 11.2% in the PCI arm vs 10.2% in the CABG arm. The hazard ratio was 1.10, and 95% CIs ranged from 0.91 to 1.32. The P value of .33 suggested that if there was no difference in the two treatments, this would not be surprising data. (No difference, right?)

But then they reported the Bayesian analysis, which revealed that the probability of any mortality benefit of CABG over PCI was 85.7% and the probability that the benefit of CABG was 1% or greater in absolute terms was 49%. (Yes, there is a likely difference, in favor of CABG.)

I cite this as a major story because Bayesian interpretations of trials make more sense for clinicians. Professor Jay Brophy from McGill University published a Bayesian interpretation of the EXCEL trial in 2020, and now that Harvard researchers have also embraced it, I hope this serves as the spark to make Bayesian interpretation of evidence more standard.

Too often in medicine, we favor therapies not because of evidence but for what John Hampton called therapeutic fashion . A practice that makes sense and "has always been done" can easily become ensconced into standard of careeven if the supporting evidence is weak.

Therapeutic hypothermia for survivors of cardiac arrest is such a practice. Cooling makes sense and has been codified for nearly two decades into guidelines. But the evidence stemmed from small biased studies done in 2002. In 2013, a larger trial demonstrated no benefit from cooling, yet the practice still did not change.

In 2021, the New England Journal of Medicine published the international TTM2 trial, which randomly assigned 1900 patients to cooling vs normothermia and found no difference in mortality or neurologic outcomes, as well as a 45% higher rate of adverse events in the cooling arm.

The message of this story transcends the treatment of survivors of cardiac arrest. Yes, it's good to know cooling doesn't benefit these patients. But the larger message for our profession is the danger of hubris and the power of the randomized controlled trial to guide medical practice.

(I refer you back to #4 and the U.S. registry of percutaneous LAA closure. Imagine if even a fraction of those 49,000 patients had been randomly assigned in a proper trial. We'd know whether this procedure actually works.)

I appreciate the support and feedback in 2021. As always, feel free to disagree with my choices, and share your top stories in the comments.

John Mandrola practices cardiac electrophysiology in Louisville, Kentucky, and is a writer and podcaster for Medscape. He espouses a conservative approach to medical practice. He participates in clinical research and writes often about the state of medical evidence.

Follow John Mandrola on Twitter

Follow theheart.org | Medscape Cardiology on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube

View original post here:
Mandrola's Top 10 Cardiology Stories of 2021 - Medscape