COMMENTARY

Jul 8, 2022 This Week in Cardiology Podcast

John M. Mandrola, MD

Disclosures

July 08, 2022

Please note that the text below is not a full transcript and has not been copyedited. For more insight and commentary on these stories, subscribe to the This Week in Cardiology podcast on Apple Podcasts, Spotify, or your preferred podcast provider. This podcast is intended for healthcare professionals only.

In This Week’s Podcast

For the week ending July 8, 2022, John Mandrola, MD comments on the following news and features stories.

IPE Benefits in REDUCE-IT Revisited

Circulation published an extremely provocative analysis from the 2018 REDUCE-IT trial of icosapent ethyl (which I will call IPE from now on, not Vascepa) vs a mineral oil placebo in statin-treated patients with high triglycerides (TG) and high CV risk. Recall that REDUCE-IT shocked the cardiology community. Over nearly 5 years of follow-up, MACE events, such as CV death, MI , stroke, unstable angina or revascularization, occurred in 17.2% of those in the IPE group compared with 22% of those in the mineral oil arm. In cardiology trials, this nearly 5% absolute risk reduction is considered a massive effect size. And in relative terms, the 25% relative risk reduction was highly statistically significant.

These results stunned because of the large effect size and statistical robustness. The second surprising feature was that these patients were already on statin drugs—a therapy known to reduce future risk of cardiac events. And most surprising of all, was that REDUCE-IT was an outlier; no other fish oil trials had come close to such positive results.

But closer inspection of the data muddied the topline results. The high-dose of fish oil led to only modest reductions in levels of triglycerides. That’s curious because reductions in TG was believed to be one of the main ways eicosapentaenoic acid (EPA) confers benefits. Experts countered these concerns by arguing, rightly, that fish oil may work through other means.

The more pressing concern was the placebo or mineral arm. Patients in this arm sustained significant increases in levels of LDL-C, apolipoprotein B, and high-sensitivity C-reactive protein—all markers of increased risk for cardiac events. This led to an existential question regarding REDUCE-IT: did IPE reduce cardiac events, or did the mineral oil placebo cause an increase in events? This debate only intensified two years later when JAMA published the STRENGTH TRIAL of an EPA-DHA supplement vs placebo in a similar high-risk statin-treated population. In fact, STRENGTH was stopped due to futility. But did it fail because EPA and docosahexaenoic acid (DHA) don’t work, or did it fail because their placebo was corn oil and there were no increases in atherogenic properties with it.

About the same time as STRENGTH came out, a separate group of researchers set out to study plaque characteristics in patients like those enrolled in REDUCE-IT and treated with IPE or mineral oil.

EHJ published the EVAPORATE Trial, which concluded that IPE resulted in plaque regression compared with placebo. However, EVAPORATE also reported a doubling of plaque volume in the mineral oil group. So, was IPE great or mineral oil terrible?

Now, finally, to the latest Circulation paper, led by inflammation expert Dr Paul Ridker. This paper took advantage of the fact that patients in REDUCE-IT had stored blood samples that Ridker’s lab could analyze for important biomarkers and compounds on the inflammatory cascade-- interleukin-1β, interleukin-6, hs-CRP, oxidized LDL-C, homocysteine, lipoprotein(a), and lipoprotein-associated phospholipase A2 (Lp-PLA2). These were available at baseline and at 24 months in both groups.

At baseline there were no differences—as you’d expect.

In the IPE group, there were no significant changes over time. But. The levels of biomarkers associated with atherosclerosis increased significantly over time among those allocated to mineral oil treatment.

The conclusions were that “The effect of these findings on interpretation of the overall risk reductions in clinical events observed within REDUCE-IT is uncertain.”

Well, that is an extremely cautious conclusion. I’d say we already knew that mineral oil increased LDL-C and hs-CRP, now we know, thanks to the authors, that mineral oil increases all kinds of atherogenic compounds.

Comments

I side much closer to the Steve Nissen opinion that it isn’t the EPA that drove the results of REDUCE-IT, it was the harm from the control arm. Proponents will counter this with modeling studies that say the amount of increased harm from mineral oil only reduces the benefit from IPE, it does not eliminate it.

This story may go down as the absolute best example of the importance of considering the control arm in trial interpretation. I would also suggest the possibility that practicing doctors knew this. Uptake of IPE was super weak, despite the massive risk reduction in REDUCE-IT.

One possibility is that pre-authorizations, cost, and pill burden kept uptake of IPE low, but another is that docs sensed the nuances of the seminal trial.

Finally, two more points: one is that this is a shining example of the importance of having not one but two regulatory trials before approval of a new drug. Millions, perhaps billions of dollars, have been spent on EPA. What if it doesn’t reduce outcomes? Second...and I’ve said this before, we need more open data.

Take EVAPORATE, if there was actually a doubling of plaque in a matter of a year in the mineral oil group, these patients would be completely occluded in short order. Major papers such as this should have to open their data and images for independent review.

US Heart Health

Circulation published an AHA president’s update on CV health. The AHA had a checklist of 7 items, called Life’s Simple 7—diet, exercise, smoking, BMI, blood lipids, blood glucose, and BP. That checklist has now been renamed Life’s Essential 8. The new addition is sleep duration. They score each of these using an algorithm from 0-100. Along with that new document, AHA also published an observational study using NHANES data to assess the status of CV health in US adults and children from 2013-2018.

The news was not good. And keep in mind it truncated at 2018. So whatever bad news I tell you, it’s fair to estimate that CV health worsened during the pandemic years. Because CV health has deteriorated in about 9 out of every 10 patients I see in clinic.

The study, led by AHA president Donald Lloyd-Jones, looked at 23k participants in the longitudinal NHANES database aged 2-79. For each participant they calculated an overall CV health score from 0 to 100, as well as the score on each of the Life’s essential 8.

80% of adults scored at a low or moderate level for CV health.

Only 0.45% of adults had a perfect score of 100; 20% had high CV health (score of 80+).

Women had better CV health than men.

In general, adults scored lowest in the areas of diet, physical activity, and BMI.

Children's diet scores were low, at an average of 40.6.

Among children ages 2 to 19 years, only (2.2%) had optimal CV health scores of 100. The proportion with high CV health declined markedly with increasing age: 56.5% of 2-5 year old children had high scores, compared with 33.5% of 6-11 year olds and 14.0% of 12-19 year olds.

Again, I reiterate, as bad as these numbers are, I will bet you two espressos that US CV health declined in more recent years—especially children who were kept out of schools and school activities.

Comments

Why is this important? A month ago, on the June 10 #TWICPodcast, I discussed a provocative study in JAMA-Cardiology that found that expansion of Medicaid led to more healthcare consumption among low-income adults in those states, but no change in CV risk factor management. And I noted that this sort of data was not an outlier. Three other major studies found the same: more access to healthcare led to more consumption of healthcare but little or no differences in outcomes.

Health likely stems more from matters upstream from clinicians.

To me this AHA data bolsters that premiseChildren should be beacons of CV health, yet, this data show that as an American toddler progress through childhood their heart health declines.

That’s critical because atherosclerosis is a slow process and the earlier it begins the more likely it is to strike in middle age.

Preventing heart disease means changing the norms of society and culture. Sure, clinicians can lead by example, by, say, riding our bikes to work and staying lean, but to really influence American heart health will require changes in societal norms.

Pharmacist Prescribing

This week the FDA decreed that pharmacists could prescribe nirmatrelvir and ritonavir for appropriate patients.

The FDA statement says that patients need to bring medical records showing recent kidney and liver function tests and a list of their medications because the antiviral has many drug-drug interactions.

Pharmacists should refer patients if they don’t know the labs or the patient’s meds or if there is a contra-indication with another drug. This could be significant because the patients most likely to benefit from the drug are higher risk patients who may be on meds that have interactions (eg the anticoagulant apixaban).

Here is why I highlight this issue on a cardiology podcast: I’ve seen and heard a fair amount of guff about this coming from American docs.

While I am sympathetic as to why FDA gets to make such a decree, almost by fiat, I want to remind listeners that just north of our border, in Canada, pharmacists have prescribing privileges. So, it may not be “normal” here, but it is in other places. And I have not heard about safety issues from my Canadian friends.

The issue with this antiviral is that it is best started soon after symptoms and a positive test. And in the US, it’s hard to get access to docs that fast.

International listeners may be perplexed to hear that, but unless you have a concierge doc, whom you pay a retainer fee, or you are very lucky, most regular Americans struggle to see a non-emergency medicine-doc urgently.

As listeners know, I am drawn to alternative care models—because our current model overspends and underdelivers. For instance, I’ve embraced the notion that ON AVERAGE advanced practice clinicians can match the level of outcomes of physicians

Most health systems embrace, letting an APC who may have a mere 2 years of training after a nursing degree, prescribe all manner of medicines. Why should we have a problem with a pharmacist prescribing a 5-day course of an anti-viral? The logic is inconsistent. And inconsistent logic really bothers me.

Finally – would this not be a golden opportunity to study different care models with cluster randomized controlled trials. Randomize parts of a hospital system, or part of a city and study a) patient satisfaction, and b) outcomes or safety.

Statin Eligibility

JAMA-Cardiology published a provocative study from Danish authors comparing ‘statin eligibility’ for primary prevention of the newest ESC guidelines.

And as I will try to explain...this study is unnecessarily complicated.

As I understand it. The ESC guides statin use based on matching the intensity of treatment with absolute risk. This I have come to believe may be a dubious principle, but let’s accept it for now.

If statins reduce risk by 25% then 25% of a higher absolute risk is a bigger number than a 25% reduction of a smaller absolute risk.

In the past, ESC used a risk score that included only CV death. It was called EuroSCORE 1. If you believe MI and Stroke are important outcomes to prevent, such a score will underestimate risk. So they came up with a new score –called EuroSCORE2—which gets closer to our pooled cohort equations or the UK NICE-QRISK for estimating absolute risk. One other important thing about EuroSCORE2 is that is geography specific. Western European countries such as Denmark, Norway, Netherlands, Belgium, the UK, etc are considered low risk by WHO because they have low CVD rates. There is a low-risk SCORE-2. I told you this was complicated.

The question that the authors of this paper had is how does the new risk score perform relative to the US and UK scores. They used the low-risk version of SCORE-2

They looked at a large cohort study in Denmark –66k people who were free of heart disease, diabetes and CKD and were not on statins. The follow up was nearly a decade. This is possible in Denmark because of their excellent health database.

The findings:

  • SCORE-2 correlated well with the other models.

  • SCORE 2 showed the best goodness of fit while having similar discriminatory performance as the other models.

  • SCORE 2 dramatically reduced Class 1 recommendations for statin use compared with the US and UK scores.

The updated age-specific recommendations dramatically reduced those with a class I indication for statins to only 4% with low sensitivity for detecting future ASCVD events. Notably, less than 1% of women met the new 2021 European-ESC treatment criteria.

The old ESC guidelines, and the US and UK guidelines provided class 1 indications for 20-34% of individuals.

The editorialists agree with the tone of the authors that the new ESC guidelines for primary prevention is not good—because so many fewer patients would have class 1 indications.

Their most compelling point comes in the problems with using geography to influence risk. For instance, Denmark has very low CV death rates and high statin use. And they argue that high statin use is one of the causal factors in producing this low CV event rate—though, having recently been in Denmark, I suspect their low risk has a lot more to do with their culture—say the notion of jante loven, and the fact that they ride their bikes places instead of using SUVs.

Comments

I think all of this is wrong-headed. First of all, statin thresholds ought to be in the eye of the beholder, not doctors. We use anticoagulation in AF to reduce a 2.2% annual risk of stroke to 1%, that’s a tiny reduction in absolute terms, and anticoagulants are higher risk than statins.

So who’s to say there is some magic 10-year risk that warrants statins.

People feel differently about this. Some would like to minimize cardiac risk to the lowest level possible and don’t mind taking a daily tablet.

Others hate pills and worry far more about cancer than an MI. Cardiologists often forget that there are many worse causes of death than heart disease. I know this because I am married to a palliative care physician.

The point is that statin use ought to be a personal decision. And guideline thresholds are dumb.

The other thing I really wonder about is whether we are wrong about this focus on absolute risk.

What if Brian Ference and the Mendelian randomization colleagues are right, and the key to maximizing benefit is the duration of exposure to LDL-C. This would argue for much earlier exposure to statins—regardless of a risk calculator. I know—there are unknown unknowns about statins because the trials are short in duration, and CV prevention is measured in decades. I will close with the comment that the point of preventive therapies is living long enough to become higher risk. The potential gains from preventive therapies may actually be greater when used at younger ages and at lower risks for longer durations.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....