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, download the Medscape app or subscribe 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 September 29, 2023, John Mandrola, MD comments on the following news and features stories.
First, some brief reader feedback. Matt Gross left a comment reminding me that I misspoke regarding the alirocumab trial called ODYSSEY OUTCOMES. I said that cardiovascular (CV) deaths and overall deaths were similar but in fact, he was right that overall death was lower in the alirocumab arm — 3.5% vs 4.1%. I also agree with Matt that this was likely spurious as there was no significant difference in CV death or death from coronary heart disease.
I would also add that overall death in the evolocumab trial called FOURIER, overall death rates were slightly higher in the active arm – 3.2% vs 3.1%. Bottom line: the PCSK9 inhibitor drugs reduce LDL cholesterol on top of statins and induce modest reductions of non-fatal CV events.
Thanks for listening so closely.
I also received a number of emails regarding my frustration over misuse of coronary artery calcium (CAC) scans. While the general consensus was agreement with me, one message I heard was that tests are just tests. They are not bad or good; what is good or bad is their use and misuse.
The blame for overuse and misuse of CAC is on the human clinicians. I remember one of my mentors, Bill Miles, reminding me of this when I was railing against the wearable cardiac defibrillator. Bill said the test or device or treatment has no properties, it is the clinicians who deserve the blame for misuse.
So I re-iterate my plea to the key opinion leaders to put out stronger statements regarding the proper use of CAC as a risk prediction enhancer for preventive therapies and not the beginning of a downstream spiral of cardiac testing.
Thanks again for the thoughtful feedback. Please keep it coming.
One more announcement. I am broadcasting from the wooded hills northeast of Oslo, Norway. We have had the first day of the 7th annual meeting of the Norwegian AF Research Network. I am honored to speak, and impressed by the basic, epidemiologic, and clinical research going on here.
Thanks to leaders Drs. Ingrid Christopherson and Arnljot Tveit for the kind invitation. It is exciting to be in Norway for the first time.
The journal Europace published a study from a group at Albert Einstein in New York comparing outcomes after standard single lead transvenous ventricular (TV) pacing and leadless pacing (LPM). It got a lot of attention on X, formerly Twitter, among some high profile people.
Leadless pacing is now a mature technology and the authors point out the lack of randomized controlled trials (RCTs) comparing leadless to standard transvenous (TV) pacing. The investigators attempted to answer the question of which device leads to better outcomes.
After this study, they concluded that there was a much higher rate of in-hospital mortality and complications with LPM vs TV-VVI pacing. They then say that this may have been related to higher co-morbidities in the LPM arm.
They are surely correct, so after I tell you about the study, I will ask a question.
The authors decided to query the national inpatient data-base from 2016 to 2019 to assess outcomes after LPM vs TV-VVI.
They then used ICD-10 codes to count complications.
This query led to large sample sizes: more than 27,000 patients had TV-VVI and nearly 8000 had LPM. Average age was 81 years in the TV group and 77 years in the LPM group.
Table 1, patient characteristics shows the main flaw in this study — there were vast differences in co-morbidities, with the LPM group having far greater numbers of co-morbid conditions.
Of course, the authors did propensity matching, a technique that attempts to find similar groups. But still these are retrospective non-random comparisons.
They report much worse outcomes in the LPM group:
Higher in-patient mortality; odds ratio (OR) 1.63.
More vascular complications; OR 7.5.
More venous thrombo-embolism; OR 3.6.
More cardiac complications; OR 1.8.
Comments. This is a problematic line of research. LPM and TV-VVI are clearly used on different types of patients. LPM are used more often in quite ill people; for instance, in patients who are at risk of infection, such as patients with end stage renal disease, or recurrent bacteremia. Having those conditions puts a patient at risk of higher rates of bad outcomes.
Adjustment can only do so much to balance co-variates. And the outsized rates of worse outcomes strongly suggest that it was not the device causing these worse outcomes, but the sicker patients.
My question, therefore, is why conduct and publish this study when it is so clearly affected by baseline differences?
What’s more, in the modern era of conduction system pacing, there would be another crucial factor affecting a head -to-head LPM vs TV-VVI pacing RCT. That is, patients in the TV-VVI arm would also benefit from conduction system pacing and the lack of dys-synchrony that comes with that.
Might I suggest something controversial? Since LPM is so often used as niche device, perhaps we do not need a randomized trial vs standard pacing. Let me know what you think.
Wow — I have a lot to say here. I will keep it brief because it affects primarily US cardiologists.
Last week I mentioned that the American College of Cardiology (ACC), Heart Failure Society, Society for Cardiac Angiography, and Heart Rhythm Society have begun talks to create their own dedicated CV medicine board certification.
Steve Stiles covered the story, and it is the most popular article on the site.
This is big news because this new independent, self-governed entity would supplant the monopoly held by the American Board of Internal Medicine (ABIM).
The oversimplified background for all listeners is that certification of internal medicine physicians and those in specialties such as cardiology has long gone through the ABIM which is overseen by a larger entity called the American Board of Medical Specialties (ABMS).
The problem — from the perspective of practicing doctors — is that certification once met passing one test after training, and then keeping up with routine CME.
Then ABIM — without representation from practicing docs — said, nope, now you need to recertify every 10 years.
Then that was not enough, ABIM added something called maintenance of certification (MOC), which requires frequent assessments and more fees.
Another problem, cardiologists learned about things like rent-seeking and regulatory capture, because much of ABIM power came from the fact that it was written into law in the Affordable Care Act.
And…professional societies made money from the ABIM by selling test materials.
I never thought anything would change, but enough doctors spoke publicly, mostly through social media channels, that professional societies felt threatened enough to embark on change.
It remains to be seen what these new board requirements will be, but it is a start.
One of the main lessons is that social media offers new ways of bringing change.
Since Steve wrote his review, dated September 22, we have learned that the CEO of ABIM, Richard Baron, is retiring. And yesterday, the American Society of Hematology, sent similar letters to ABIM suggesting that they too were interested in independent board certification assessments.
It looks like I might have been wrong about this situation. I had thought it was hopeless. That we doctors had no collective power against entrenched bureaucracy. We shall see how it goes. I am also watching the movement in the UK to affect change in the National Health Service.
There is great hope for positive change if only our profession could put aside personal goals for the greater good. Professional sports team do this, but I’ve yet to see doctors do it.
SGLT2 Inhibitors for CKD
This week, the US Food and Drug Administration approved the use of empagliflozin for a new indication, that is, treating adults with chronic kidney disease (CKD) who are at risk for progression, even in the absence of type 2 diabetes (T2D) or heart failure (HF). Just CKD.
This is the second approval for prevention of CKD progression; dapagliflozin received that approval back in 2021.
The approval was based on EMPA-Kidney, published 11 months ago in the New England Journal of Medicine (NEJM).
In brief, EMPA-Kidney randomly assigned 6600 patients who had CKD as measured with two criteria:
An estimated glomerular filtration rate (eGFR) of at least 20 but less than 45 ml/ min/ 1.73m2 of body surface area or
An eGFR of at least 45 but less than 90 ml/ min/ 1.73m2 with a urinary albumin-to-creatinine ratio (with albumin measured in milligrams and creatinine measured in grams) of at least 200.
At 2 years, empagliflozin reduced the primary composite endpoint of progression of kidney disease or CV death by a significant 28% relative to placebo.
This is now the fourth FDA -approved indication for empagliflozin, which was previously approved for:
Reducing CV death and hospitalizations for HF in patients with HF regardless of ejection fraction;
Reducing CV death in adults with T2D who also have known CV disease;
Reducing blood sugar (along with diet and exercise) in adults and children older than 10 years old who have T2D.
Comments. Cardiologists are clearly prescribing these drugs for HF. But a lot of our non-HF patients, those with coronary artery disease (CAD), atrial fibrillation (AF), and premature ventricular contractions also have CKD and T2D and we ought to be on the lookout for patients who are not on these drugs.
Thing is, a lot of times, I have contacted the primary internist about starting an SGLT2 inhibitor for CKD or T2D and the response is that they have already tried using them, and the patient could not afford the drug. So cost remains an issue.
I sit on the peer review committee, and we have seen a number of cases of non-hyperglycemic ketoacidosis due to SGLT2 inhibitors.
The problem is patients come in with generalized symptoms and have super low bicarb levels, high anion gaps, recognition is slow, and treatment can be delayed.
So I guess my main message is for increased alertness to the low bicarb and high anion gaps and earlier recognition of this issue.
Finally, the increased issue with this complication speaks to the difference between trials and real-life practice. Metabolic acidosis due to SGLT2 inhibitors may not have been a major problem in trials, because trials have research nurses and careful follow-up.
That’s a different environment than the real-world practice. Practicing docs will get the hang of it, eventually, but right now, be on alert for keto-acidosis from SGLT2 inhibitors.
Let’s talk about diagnosis of myocardial infarction (MI) after cardiac surgery. What is super weird is that modern technology has made things harder, not easier, when it comes to diagnosis.
When I started cardiology, it was simpler to diagnosis MI after cardiac surgery. You had to have serious focal ST-T changes on an ECG, or major new echo wall motion abnormalities. Enzymes might help, but creatine kinase (CK)-MB was always high.
And it is important to make this diagnosis because rapid re-operation or perhaps percutaneous coronary intervention (PCI) might help fix graft failure.
The problem is obvious, right? Surgical trauma causes release of cardiac troponins. The question is how much is acceptable. And how much is enough to predict mortality or initiate acute treatment decisions.
Now, though, the advent of more sensitive troponin assays has complicated the matter, and the complications transcend clinical practice and get into trial interpretation, because a lot of the controversy surrounding EXCEL (coronary artery bypass graft (CABG) vs PCI in left main coronary artery disease) turned on the definition of MI in the two groups.
The Journal of the ACC has published an important effort from a group in Innsbrook Austria, first author Leo Polzl.
In a single center, with consecutive group of more than 8200 patients who had cardiac surgery, the authors attempted to sort out new definitions of perioperative MI after cardiac surgery using high sensitivity cardiac troponin T (hs-cTnT) measures.
Specifically, they plotted the relationship between post-op troponin levels with 30-day mortality or 5-year mortality.
They adjusted using a EuroScor and considered thresholds for isolated CABG, aortic valve replacement AVR and other cardiac surgeries.
This is complicated work, and it’s hard to make it simple. I will try.
They measured hs-cTnT levels (Roche diagnostics) pre-op and then post op for 3 days. The highest value was then used for the comparison with 30-day and 5-year death rates.
The main figure requires closing your eyes and imaging a plot with x-axis being the troponin levels, the y-axis the hazard ratio (HR) for deaths at 30 days. The plot begins with a line going up gradually. The higher the troponin the greater the HR.
The line also includes the confidence interval (CI). And when the lower bound curve reaches a HR of something greater than 1 or then greater than 2, the authors can define a level that is so many times above the upper limits of normal (URL).
You probably won’t remember the actual numbers but here they are:
hs-cTnT value upon surgery, which was associated with an adjusted HR of more than 1.00 for death within 30 days, was 1,236 ng/L (95% CI: 1,014-1,537; 88.3 times URL) for CABG.
490 ng/L (95% CI: 363-852; 35.0 times URL) for aortic valve replacement.
The authors then tested the utility of these measurements with ROC curves. The highest predictive value of hs-cTnT was for CABG patients (AUC: 0.796; 95% CI: 0.716-0.876). Again, for ROC curves 0.5 is a coin toss and 1.00 is perfect.
Then the authors went on to try to predict the threshold to identify what is an actual peri-operative MI. They found:
The threshold hs-cTnT level for predicting 30-day mortality was 2,385 ng/L (170 times URL) for isolated CABG.
Within a multivariate regression model for the 30-day mortality, hs-cTnT levels above the cutoffs resulted in an HR of 12.56 (95% CI: 6.81-23.16; P < 0.001) for isolated CABG.
Post-op troponins in this study had little predictive value for 5-year deaths.
Comments. This is hard. An editorial-from the McMasters group accompanied this paper. It helped a bit. They cite another similar effort they called the VISION Study published in NEJM in March 2022.
VISION had the same goals of trying to sort out threshold levels of cardiac enzymes after cardiac surgery. They had more patients, 13,000. vs 8,000 in this study. VISION also had multiple centers. But they used a different hs-troponin — troponin I.
The main graphs of results were similar. X-axis were troponin levels, y-axis was HR for death at 30 days. And the plot goes up gradually and you select a troponin level associated with higher HR for death.
They found that "among patients who underwent isolated CABG the threshold troponin level, measured within 1 day after surgery, that was associated with an adjusted HR of more than 1.00 for death within 30 days was 5670 ng/L, a level 218 times the upper reference limit.
Here is one commonality – and if you take one thing to remember, it is that both studies find that the threshold troponin for correlating with higher rates of death is much much greater than the current thresholds.
In the Austrian study it is was 170 times URL and in VISION it was 218 times.
So, at minimum, in the hS-troponin era, we will need to hold off on diagnosing peri-operative MI unless there are much greater enzyme levels.
But even then, the prediction is far from perfect. Clearly, we need more work in this area to better define levels and outcomes.
For research, I wonder whether a consideration is to remove peri-operative MI or even all MI from the composite endpoints. There are so many variables — the type of surgery, the timing of the measurement, the type of assay — and even then, the level of enzyme tells us little about the cause of the perioperative MI, such as simple trauma, prolonged pump time, or graft failure.
At least until we better understand the relationship between hs-troponin releases and outcomes, I wonder about the utility of using perioperative MI in trials.
It’s funny, the better we get at biomarkers, the less valuable MI seems as a surrogate outcome measure.
People who are experts in cardiac enzymes, let me know if you think I am off base here. It’s a difficult area. I am happy to be educated.
American Heart Association 2023 Meeting
This week AHA announced their lineup of late-breaking trials, and gosh there will be big news
SELECT trial: Cardiac outcomes of semaglutide in obesity, which we know met its primary endpoint.
ORBITA-2: ORBITA 1 in more complex CAD. We shall learn more about the placebo-controlled effect on angina.
CARDIA-SSBP: A randomized crossover trial with sodium and systolic blood pressure
AZALEA-TIMI-71: The phase 2 trial of the Factor XI inhibitor abelacimab vs rivaroxaban in patients with AF
© 2023 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Sep 29, 2023 This Week in Cardiology Podcast - Medscape - Sep 29, 2023.