Paul D. Thompson, MD

This article was originally published on 

 report of a trial comparing semaglutide, a glucagon-like peptide-1 receptor agonist (GLP-1 RA), versus tirzepatide, a drug that has both GLP-1a and glucose-dependent insulinotropic polypeptide activity.

 GLP-1 RA’s stimulate insulin secretion during hyperglycemia and decrease glucagon secretion during euglycemia and hypoglycemia.GLP-1 RA’s also decrease gastric emptying thereby decreasing appetite and body weight. Tirzepatide has GLP-1 RA activity, but also stimulates insulin release and decreases glucagon levels during euglycemia and hypoglycemia via its glucose-dependent insulinotropic activity. This dual-action has theoretical advantages. The study compared semaglutide 1 mg SQ weekly, versus tirzepatide 5, 10, or 15 mg SQ weekly. Tirzepatide has not yet been FDA approved.

All 3 tirzepatide doses decreased mean hemoglobin A1c (A1c) by 2.01 to 2.3% versus 1.86% for semaglutide. Bodyweight decreased 1.9 to 5.5 kg more with tirzepatide than with semaglutide over the 40-week study. But this was not a fair comparison because STEP 2 or the Semaglutide Treatment Effect in People with Obesity study used a semaglutide dose of 2.4 mg weekly,

 and higher semaglutide doses probably would have reduced hemoglobin A1c and body weight more.

But here’s the real question. Why is a cardiologist (me) studying an article on new diabetes treatments? I remember similar thoughts about why cardiologists were interested in lipid metabolism when lovastatin was FDA approved in November 1987. And look where cardiologists are now in managing lipids using statins, PCSK9 inhibitors, ezetimibe, and bempedoic acid. Cardiologists, like all clinicians, get interested in problems they can fix easily. Metformin, sodium-glucose transport type 2 inhibitors (SGLT2i), and GLP-1 RAs (as the oral semaglutide, Rybelsus) are now all available orally. Even when using GLP1a’s as injection is not difficult. They are easy to use because they rarely cause hypoglycemia unless administered with insulin or a sulfonylurea. This means they are “smart drugs” in that their glucose-lowering effect dissipates as the glucose levels fall. All reduce A1c and reduce body weight, and the latter effect can make you a hero to your patients!

I’ll bet that our experience with LDL management, that “even lower is even better”, will also ultimately be applied to A1c management. The Fourier Trial demonstrated a 2% absolute risk reduction in cardiovascular events after only 3 years when high-risk coronary patients reduced their median LDL-C from 92 to 30 mg/dl using the PCSK-9 inhibitor, evolucamab.

 I realize that available diabetes trials show no difference in outcomes with A1c’s above or below 7%, but these trials used insulin to achieve the A1c levels, and insulin causes hypoglycemia and has many atherogenic properties. Furthermore, someone with “borderline diabetes” has already lost 50% of his/her insulin-secreting capacity. The Diabetes Prevention Program demonstrated that treating patients with prediabetes, defined as an oral glucose tolerance 2-hour value above 140 mg/dl, but less than 200 mg/dl, decreased the onset of diabetes by 31%.

 Hyperglycemia not only glycosylates hemoglobin, but also glycosylates proteins in the LDL particle making LDL more inflammatory and atherogenic. In the Early ACS Trial, among 879 patients with non-ST segment elevation MIs (NSTEMI), 32.5% had known diabetes, 12.2% had undiagnosed diabetes but fasting glucose values at or above 126 mg/dL or an A1c greater than 6.5%, and 8% had prediabetes diagnosed by a glucose of 110 to 126 mg/dL. 5Only 44.4% had normal glucose metabolism. Prediabetes is dangerous. It leads to diabetes and causes cardiac events along the way.

Maybe it’s time for cardiologists to know more, do more, and be more aggressive about treating diabetes, borderline or otherwise.

Frias JP, Davies MJ, Rosenstock J, et al. Tirzepatide versus Semaglutide Once Weekly in Patients with Type 2 Diabetes. N Engl J Med 2021;385:503-15.

Davies M, Faerch L, Jeppesen OK, et al. Semaglutide 2.4 mg once a week in adults with overweight or obesity, and type 2 diabetes (STEP 2): a randomised, double-blind, double-dummy, placebo-controlled, phase 3 trial. Lancet 2021;397:971-84.

Sabatine MS, Giugliano RP, Keech AC, et al. Evolocumab and Clinical Outcomes in Patients with Cardiovascular Disease. N Engl J Med 2017;376:1713-22.

Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393-403.

Giraldez RR, Clare RM, Lopes RD, et al. Prevalence and clinical outcomes of undiagnosed diabetes mellitus and prediabetes among patients with high-risk non-ST-segment elevation acute coronary syndrome. Am Heart J 2013;165:918,925.e2.

Articles

Editorial advisory board member Dr. Paul Thompson reflects on how advances in diabetes management have impacted cardiovascular medicine and why staying abreast of these advances is essential to being a cardiologist in 2021. 

Paul Thompson, MD: Why Cardiologists Need to Be Diabetologists

Somewhere during my medical school training, someone told me that if you really want to improve medical care, you should be a good teacher because being a good teacher magnifies your effect. Dr. Thomas Lee’s book on Eugene Braunwald

states that when Braunwald took over as Chief of Medicine at the Brigham, he considered getting the right residency director as one of his most important task.

His reasoning? Having a strong residency director would create a competitive residency that would attracted the best residents, who would provide better care.The best of the best residents could subsequently be recruited to join the faculty. Having a strong residency program also makes an institution more attractive to attendings who like to teach.

One of the most important effects of teaching, is that the teacher gets smarter. I always say that “Teaching Teaches the Teacher” or T4. Dr. Braunwald also knew that his trainees would go elsewhere across the country and if the residents were well-trained, they would enhance the Brigham’s reputation. It’s an upward spiral. Dr. Braunwald recruited Sheldon Wolff, who became a beloved leader of the Brigham residency program.

All of this makes me wonder if we reward clinical teachers well enough. Good teachers can make a place better. I am not referring only to teaching medical students and trainees. Constantly teaching nurses and other clinicians if done with respect and absolutely no condescension can improve clinical quality. A teaching environment also attracts job candidates to such an institution.

I once worked where we had lots of applicants for our advanced practice nurse practitioner (APRN) positions, despite not paying top dollar. I asked one of our APRN’s why she worked there when she could get more pay at just about any other hospital. She replied that her plan was to work there for a couple of years because the teaching/training was so good. She would then be more desirable at other places. I was disappointed when that APRN left, but had the satisfaction of knowing that what she learned from me would make someplace else, and its care, better.

We should use every opportunity to teach. It makes us and those we work with smarter. I start my clinical consult notes with, “Thank you for asking me to evaluate Mr. Jones. I made the following recommendations for the following reasons”.The goal is to make my consult notes educational, but does it work? I think so. When I first started the lipid management clinic at Hartford Hospital in 1997 approximately 5% of patients sent to me for hypercholesterolemia were hypothyroid. So, my first line would be: “I started Mr. Jones on thyroxine 25 mcg daily because his TSH is elevated indicating that he is hypothyroid. Hypothyroidism decreased LDL receptor activity, which reduces the clearance of LDL-cholesterol and increases LDL blood levels.” 

After 24 years of such letters, I almost never have a patient sent to me with hyperlipidemia who is hypothyroid. Are my notes totally responsible? I doubt it, but they may be partly responsible.

You do not have to practice at a teaching hospital to teach. Teaching the nurses and others around you is critically important. It makes their jobs more interesting, but also means that they can help you avoid oversights. Smart staff can also make suggestions to other doctors in the practice to help improve overall medical care.

Dr. Braunwald got it right, as he so often has. Being a good teacher and using every possible opportunity to teach creates an upward spiral for you, your clinical care and knowledge, and your institution. Good teachers should get more credit for the unmeasurable good they do to improve medical care.

1. Lee TH. Eugene Braunwald and the Rise of Modern Medicine.Harvard University Press, Cambridge, MA. 2013

Dr. Paul Thompson reflects on his experiences as a practicing cardiologist, including having the ability to impact early-career clinicians and what he gains when presented with the opportunity to teach other providers. 

Teaching Teaches the Teacher, Where Academia and Clinical Medicine Meet

Over the last 6 months, I have had two patients with more than a heart and lungs in their chest. The first was a retired teacher who developed dyspnea mowing his lawn. It sounded cardiac to me, but his echocardiogram and exercise stress test were normal. I ordered a chest x-ray because, years ago, I had a patient with similar symptoms who turned out to have a paralyzed right hemidiaphragm. 

I was embarrassed by missing that prior patient and, consequently, very proud of myself when the present patient’s chest x-ray showed an elevated right hemidiaphragm. I shipped the patient off to a pulmonologist, who did a radiographic cine sniff test that showed normally moving hemidiaphragms bilaterally, but a very large hiatal hernia. The patient is scheduled for closed hiatal hernia repair because the pulmonologist is convinced that this is the cause of the patient’s dyspnea.

I also recently saw a distance runner with palpitations. He had had palpitations (and gastro-esophageal reflux) for several years but a recent prolonged episode of palpitations sent him to the emergency department (ED). I ordered an event monitor that showed that 10% of his beats were PVCs. His echocardiogram and left ventricular function were normal, but his chest x-ray from the ED showed a large hiatal hernia. Large hiatal hernias have been associated with ventricular ectopy, which was cured by hernia repair.

 We did not pursue hernia repair in this patient because his left ventricular ejection fraction was normal, and he and his wife declined because the patient had early dementia.

So, what’s the point? Consider doing a chest x-ray when patients have dyspnea or frequent palpitations, but no history of lung disease, and a normal cardiac workup. Sometimes there is more in the chest than just a heart and lungs, like a paralyzed hemidiaphragm or most of their stomach.

Posterior cardiac compression from a large hiatal hernia-A novel cause of ventricular tachycardia.

HeartRhythm Case Rep. 2018 May 23;4(8):362-366.

In his most recent column, Dr. Paul Thompson offers perspective on why he feels clinicians should consider performing a chest x-ray in patients with dyspnea or frequent palpitations, but no history of lung disease and a normal cardiac work-up. 

More Than A Heart and Lungs in The Chest, with Paul Thompson, MD

Cardiac non-compaction (NC) can produce a form of cardiomyopathy (CM) typified by deep invaginations of the myocardium caused by an arrest of normal myocardial morphogenesis. NCCM was considered to be rare, but is increasingly diagnosed because of the widespread use of cardiac imaging. Patients with NCCM can present with syncope and systemic emboli believed to originate from clotting in the myocardial invaginations.

Sports cardiologists have suspected that athletes, especially black athletes, have an increased prevalence of imaging evidence of NCCM, but data confirming such a relationship were lacking. De la Chica and colleagues

 evaluated the relationship between physical activity (PA) and the imaging criteria for NCCM in 705 men and women in the Spanish Progression of Early Subclinical Atherosclerosis study (PESA). PA was measured using accelerometry, the gold standard for PA assessment. LV geometry was assessed using cardiac MR.

The diagnosis of NCCM was determined using the Petersen, Jacquier, Grothoff, and Stacey clinical scoring criteria. 1PA was divided into 6 activity groups from the least (0) to the most (5) active. Some of the subjects were very active, but the population was not a population of athletes. The prevalence of NCCM using the most permissive diagnostic criteria (Petersen) was significantly higher in the most active group (30 & 14% in PA group 5 vs 0, respectively).

The prevalence of NCCN was also significantly higher in the most active group using the middle 2 diagnostic criteria (Jacquier & Grothoff), but not different between groups using the most stringent criteria (Stacey). This confirms the clinical impression that more active subjects meet the imaging criteria for NCCM more frequently than the general population. These results should help prevent clinicians from diagnosing NCCM in athletes without supporting data such as a family history of cardiomyopathy, a reduced ejection fraction, or decreased exercise capacity. I have found that athletes can have the excrescences and pits suggestive of NCCM, but that the wall thickness is normal under the “non-compacted” area. In contrast, true NCCM often has wall thinning below the truly non-compacted segment.

Black athletes appear to have more highly trabeculated left ventricles making them more vulnerable to over-diagnosis

 and seem to have the appearance of non-compaction more frequently than whites. I have seen several athletes over my career who were diagnosed with NCCM based on imaging criteria alone and restricted from sports because of the diagnosis. This is never an easy decision, but clinicians should strongly consider the possibility that non-compaction in a healthy athlete, especially a healthy black athlete, is an abnormal image, but not necessarily a disease unless there is other supporting clinical evidence.

One of the most important tenets of medicine is “Do no harm”.An important corollary is “Do not over-diagnose” because it can cause harm. 

Sports participation in non-compaction cardiomyopathy: a systematic review.

 Br J Sports Med. 2014 Oct;48(20):1466-71. 

, et.al.Association Between Left Ventricular Noncompaction and Vigorous Physical Activity.J Am Coll Cardiol. 2020 Oct 13;76(15):1723-1733.

 J. A diagnostic dilemma in non-compaction, resulting in near expulsion from the Football World Cup. Case Reports - Eur J Echocardiogr . 2011 Feb;12(2):E8.

In his latest column, Dr. Paul Thompson offers perspective on the diagnosis and potential overdiagnosis of non-compaction cardiomyopathy in athletes.

Diagnosis of Cardiac Non-Compaction in Athletes, with Paul D. Thompson, MD

I published my first article on exercise-related sudden cardiac death (SCD) in 1979.

 It described 17 men and 1 woman who died jogging or running. That doesn’t seem very novel or worthy of a JAMA publication because we all now know that vigorous exercise acutely, but transiently, increases the risk of SCD and acute myocardial infarction (AMI). But in my defense, that article was published during the jogging boom of the 1960-70s and, during the time, that Dr. Thomas Bassler, a California pathologist, maintained that running a marathon (a 42 km footrace) conferred protection against coronary artery disease (CAD) events.

Dr. Bassler had an extensive publication record based entirely on letters to the editor that cited prior letters to the editor as verification for the “Bassler Hypothesis”. So, that paper got a lot of publicity. That publicity was even greater because Jimmy Carter, then-President Jimmy Carter, nearly collapsed on September 15, 1979, while running a road race. The New York Times’ front page had a picture of the President being supported by Secret Service agents. My article was to be published on September 21, 1979, but JAMA moved it up so that Jimmy and my article shared the Times’ front page.

Not long after, I received a letter from Dr. Asher Black. Dr. Black was a practicing cardiologist in Syracuse, NY. In 1975 he had described autopsy or angiographic evidence of a “coronary artery injury” in 12 patients whose ACS or SCD occurred during exercise. 

 Dr. Black’s report preceded widespread recognition of the role of plaque rupture in ACS. He called it “Black’s Cracks in the Plaque”. We never met, but corresponded until his death. 

Later, I met Dr. Peter Libby when he was at Tufts. Dr. Libby told me that he was researching what prompted acute cardiac events, which became the focus of his life’s work. Dr. Libby subsequently summarized the characteristics of the vulnerable plaque including a lipid rich core, increased inflammatory cell activity, and a thin fibrous cap.

 The theory is that an increased lipid rich core, attracts white blood cells that make collagenase that dissolve collagen in the fibrous cap contributing to plaque rupture. Even later, I met a researcher from Dr. Renu Virmani’s lab, then at the Armed Forces Institute of Pathology. This person was presenting a poster on plaque erosion in SCD victims demonstrating that plaque rupture was not the entire story.

Most of us are more familiar with plaque rupture than plaque erosion, but a recent JAMA review by Dr. Libby and others suggests that plaque erosion as cause of acute ACS is becoming more frequent.

 That’s because widespread use of statins and other lipid-lowering medications reduce plaque cholesterol and may reduce ACS events from plaque rupture but not from plaque erosion. Flow disruption in the coronary arteries activates the innate immune toll-like receptor 2. This can lead to endothelial cell death with subsequent desquamation of the endothelial surface. This may lead to the formation of neutrophil extracellular traps (NETS) and platelet-rich coronary thromboses. As cited in the review, 30% of SCD victims and 25-44% of ACS patients have plaque erosion as the cause of their cardiac event. 4

Victims of plaque erosion are different from those with plaque disruption.4 They tend to be younger, with fewer traditional risk factors, more non-ST segment elevation MIs (NSTEMI), a lower total plaque burden, more frequent LAD involvement, and hemoglobin (hgb) values >15 g/ml. This hemoconcentration increases blood viscosity and could increase wall stress. Indeed, among 593 patients with ACS, 73% of those less than 68 years of age, with anterior wall ischemia, no diabetes, normal kidney function, an hgb greater than 15, and a NSTEMI had plaque erosion of the cause.

This is of only scientific interest now, but Dr. Libby and colleagues suggest that patients with ACS and plaque erosion may be treated medically without stenting in the future because they tend to have larger coronary lumens, more platelet rich thrombi, amenable to anti-platelet therapy, and less other atherosclerosis. Only the future will tell, but now it is becoming increasing clear that Black’s crack in the plaque is not the whole ACS story.

1. Thompson PD, Stern MP, Williams P, Duncan K, Haskell WL, Wood PD. Death during jogging or running. A study of 18 cases. JAMA 1979;242:1265-7.

2. Black A, Black MM, Gensini G. Exertion and acute coronary artery injury. Angiology 1975;26:759-83.

3. Libby P. Molecular bases of the acute coronary syndromes. Circulation 1995;91:2844-50.

4. Kolte D, Libby P, Jang IK. New Insights Into Plaque Erosion as a Mechanism of Acute Coronary Syndromes. JAMA 2021;325:1043-4.

Dr. Paul Thompson reflects on the cardiology's community journey of understanding related to coronary plaque, sudden cardiac death, and acute coronary syndromes.

Black's Crack in Coronary Plaque, with Paul D. Thompson, MD

Here is a transcript of a brief conversation I had with a colleague related to risk of statin-related tendon rupture in a patient.



 I’m aware about statin-related tendon rupture possibilities. Today, I saw a 40s-year-old lady on atorvastatin 20 mg since childhood (probable familial hypercholesterolemia) who had spontaneous ligament rupture in her hand. Per ortho, this a very rare thing, especially in someone her age. No known arthritis.



: We have written about tendon ruptures in statin-treated patients, but it is very uncommon. In 2007, we reported 267 tendon ruptures associated with statins from the FDA adverse event database and a case of bilateral biceps tendinopathy (Pulatt et al.Am J Cardiol 2007;100:152–153). Subsequently, one of my colleagues, Lane Duvall, here at Hartford Hospital, taught me that TTR amyloid can cause bilateral biceps tendon ruptures years before symptomatic cardiac amyloidosis is obvious, and this person's tendinopathy turned out to be related to TTR amyloidosis. I wrote an update of that case with Lane and Sabina Arora, who runs our amyloid program (Thompson, PD et al.Am J Cardiol. 130: 165. 2020). We also reported a 44-yr old physician with multiple tendon ruptures during 19 years of statin treatment and none during 5 years off of statins (Gowdar SD, et al. J Clin Lipidol. 14:189-191, 2019. He was referred from out-of-state. I tried to find him subsequently to make sure he did not have amyloidosis, but could not locate him. Tendon rupture with statins is theoretically possible because matrix metalloproteinase (MMP) is required for tendon repair and statins reduce some MMP’s activity.

But before blaming the statin make sure to consider other possibilities. Hypercholesterolemia alone can cause tendinopathy. In fact, I always tell our sports folks that if they see Achilles tendinitis in a young athlete to consider familial hypercholesterolemia. I did not know of tendon rupture associated with hypercholesterolemia alone, but a PubMed search using the terms “statins” and “tendon” did uncover multiple reports examining the association of various tendinopathies with hypercholesterolemia and one report of 6 spontaneous tendon ruptures in patients with familial hypercholesterolemia in Japanese (Sakai, T et al.Nihon Ronen Igakkai Zasshi, 20:138-42, 1983). 

So, it looks like both statins and hypercholesterolemia are associated with tendon rupture but if the statins are causal is theoretically possible, but unclear. Interestingly, the FDA required that bempedoic acid list tendon ruptures as a possible side effect. Bempedoic acid is a new lipid-lowering agent that inhibits citrate lyase in the mevalonate, cholesterol synthesis pathway upstream from HMG CoA reductase. I reviewed the tendon cases for Esperion, the company that makes bempedoic acid (I did it for free). The tendon issues were the typical tendon problems that happen to middle-aged people so it was hard to pin the tendinopathy on the drug.

As for your patient, I would play it safe and switch to a PCSK9, but also would think about other causes of tendinopathy like TTR amyloid.

Thanks for the question. 

In this column, Dr. Paul Thompson provides an account of a recent conversation with a colleague related to the potential for tendon issues, including tendon rupture, in a patient on statin therapy. 