The case that we discussed was atrial fibrillation after knee surgery. It’s a common post op complication that usually doesn’t mean anything serious, but in rare instances can be a harbinger of a life threatening complication.

It’s the most common type of scenario that you see in clinical practice - something important but not life threatening, where there’s no absolute right or absolute wrong, and no randomized trials to guide your decision making.

In other words, every single aspect of decision making is basically a judgment call.

I thought this was a way more interesting way of thinking about AI in medicine than how it’s often discussed, primarily because this type of ambiguity and uncertainty is really the heart of clinical medicine.

It’s basically man vs. machine for an area where there is no winner, just slightly different ways of doing things.

In the interest of comparing and contrasting human versus AI, I thought it would also be interesting to feed the transcript into Claude and see what it thought I should write as an introduction versus what I thought was the important thing to focus on.

So here is my take and then below you can see Claude’s unedited introduction.

My take: what you talk about and what you focus on is what you prioritize

I will often teach that you don’t have an infinite amount of time and an infinite amount of focus from yourself, your patients, or your trainees.

And so what you decide to prioritize is what you think is important.

In this type of scenario - post op Afib - my number 1 priority is to make sure that the patient doesn’t have Afib as a consequence of a life threatening complication like a pulmonary embolism (blood clot in the lungs), significant blood loss related to surgery, or a new infection and sepsis.

Once we’ve ruled that out, we’re basically left with a question about whether to get the person out of Afib with a procedure or whether to leave them in Afib.

I think this boils down to two major questions:

* How confident are we that the patient is truly asymptomatic?

* Would waiting to do something about the Afib comprise any part of the surgical recovery?

I landed on getting the patient out of afib with a procedure called a cardioversion because of what I knew was coming next for them - rehab after orthopedic surgery.

How someone is able to rehab after surgery is often the difference that impacts long term quality of life.

And if the patient is symptomatic because of Afib, even a little bit, it might compromise their rehab. Since you often can’t truly tell in a hospitalized patient whether there are symptoms - they’re often lying in bed, recovering from a procedure, and unable to tell the difference between the standard feeling of post op recovery and a truly symptomatic irregular heartbeat - then you’re better off doing a low risk procedure to get them out of the irregular heartbeat and remove all doubt that there is something new and cardiac that will hold up their rehab.

Your mileage may vary whether I’m correct - but my thought was that once we’ve ruled out something life threatening, we should think about what might improve quality of life next.

And like many places in medicine, there is no absolute right and absolute wrong. Most of my days compromise a million judgment calls that I’m not certain about. And if AI can help me make more of those a bit better, I am open to its feedback.

Claude’s take: "Can" vs. "Should": What the AI Missed on a Routine AFib Case

I just did something a little different for Core IM’s new Bread & Butter series. I worked through a real post-op AFib case out loud — and then we compared my reasoning, step by step, against what OpenEvidence recommended for the exact same patient. Gurpreet Dhaliwal refereed the human-vs-AI matchup.

The case sounds simple. A 63-year-old man with diabetes and hypertension gets a knee replacement, and afterward he goes into AFib with RVR. The kind of consult you get paged about constantly. But the longer you sit with it, the more you notice how much of what we call “standard management” is really just convention — handed down and codified, often without the level of evidence any of us would want if it were our own family in that bed.

A few things I keep coming back to:

“Can be discharged” and “should be discharged” are not the same sentence. The AI told my resident the patient could go home once he was rate-controlled. Totally defensible. Medico-legally fine. But that’s the floor, not the ceiling — and the gap between the minimum acceptable care and the care you’d actually want for someone you love is where most of the real medicine lives.

CHA₂DS₂-VASc isn’t measuring the left atrial appendage. We get overly concrete about why AFib causes stroke. The clot-in-the-appendage story is one mechanism, but look at what’s actually in the score: nothing about the appendage, nothing about atrial size or emptying velocity. It’s essentially a measure of how much atherosclerotic disease a patient has. Not everyone with AFib who strokes is even in AFib when they stroke. Don’t let the score become the whole conversation.

Sometimes the best stroke-prevention drug isn’t a blood thinner. If this man’s AFib had been a five-minute self-limited run, he’s arguably better served by a GLP-1 than an anticoagulant when it comes to lowering his stroke risk. Stroke prevention in AFib is a whole-patient problem — sleep apnea, lipids, coronary calcium, blood pressure — not just a binary decision about whether to start Eliquis.

The AI never asked whether he was symptomatic. I did, on purpose — and it matters enormously, because plenty of “asymptomatic” patients get cardioverted and only then realize they’d been feeling lousy for weeks. Our job is to help people live longer and live better. You can’t do the “better” half if you never ask how they feel.

What I loved about this format is the thing Gurpreet kept circling back to: uncertainty isn’t a bug in medicine, it’s the job. If there were one provably right answer, you wouldn’t need a doctor. The AI gave confident, defensible advice — and ask it the same question twice and you’ll get two different defensible answers. Judgment is what fills that gap.

🎧 Listen here: Post Op Afib Bread and Butter Series

If it resonates, do me a favor and send it to one colleague — and I’d love to hear how you’d have managed this one in the comments.

There are more than a billion people walking around right now with elevated blood pressure, and it’s the driver of a huge share of strokes, heart attacks, kidney failure, heart failure, and dementia.

Doctors call it the silent killer. I think we should call it the boring killer, because it’s more accurate. Nobody wants to talk about blood pressure because it doesn’t generate scary headlines or sell supplements.

It’s just relentlessly, unglamorously important.

The full episode is here. I wanted to summarize a bit of what we got into, in case you prefer text over video.

Blood pressure is a moving target, and too often we treat it like it’s a fixed one

Your blood pressure is supposed to fluctuate.

It rises with exercise, with stress, with a salty dinner, with a bad night of sleep, with an argument on the drive to the doctor’s office. That’s normal physiology, not pathology.

And yet I think that there’s something truly absurd about how we operationalize this: we take a measurement that varies constantly, check it once in a clinic, and then make a decision about a daily pill that follows someone around in their chart forever. If your blood pressure is high in the office, the answer is to avoid overreacting. You almost always want to get more data.

Nobody is going to stroke out this afternoon because their reading was 145/90 today.

A home blood pressure cuff costs about $40, and the expensive ones aren’t more accurate, they just have Bluetooth.

High blood pressure is not a moral failing

This came up repeatedly, and I think it’s one of the most important ideas in the episode.

There’s a strong cultural current right now that treats every biometric as a referendum on your discipline. But I see plenty of patients who do everything right - they exercise daily, they barely drink, they sleep well, they even meditate - and their blood pressure is still elevated.

Age, genetics, and the natural stiffening of arteries over time are not character flaws. They’re what being a mortal human looks like. Ending up on a medication isn’t a defeat, and it isn’t a life sentence either.

If your circumstances change - you lose weight, the stressful season ends, the sleep apnea gets treated - the plan can change too.

The trap of the indefinite trial of lifestyle

It’s easy for doctors and patients to agree on “let’s try diet, exercise, and stress management changes for three months” - and then three months becomes 6, becomes 9, and suddenly it’s 3 years later and the blood pressure has been elevated the whole time.

Risk from blood pressure accumulates based on how high it is and how long it stays there. It’s not an emergency, but it’s also not something to sit on indefinitely, especially when the medications we have are cheap, well tolerated, and effective.

How I actually choose medications

We walked through the real-world pharmacology: why I reach for an ARB first in patients with diabetes, why amlodipine is a great choice for people who hate getting blood drawn (no lab monitoring required), and my standing joke that amlodipine’s three doses (of 2.5mg, 5mg, and 10mg) that could be renamed placebo, blood pressure treatment, and leg swelling.

We talked about why thiazide diuretics work well on paper but create problems in real life when a patient has a long commute or a prostate that already wakes them up three times a night.

The mundane side effects are the ones that determine whether people actually take their medications, and pretending otherwise is how treatment fails.

Why beta blockers fell from grace

If you’ve been reading this newsletter for a while, you know this is a recurring theme for me.

Beta blockers lower the number on the cuff without adequately lowering central aortic pressure, something that should be called pseudonormotension.

Blood pressure looks controlled while the brain and kidneys are still experiencing elevated pressures, which is probably why beta blockers consistently underperform other antihypertensives for stroke prevention.

There are absolutely patients who should be on them, like heart failure with reduced ejection fraction, heart rate control for afib, but for garden-variety hypertension, they shouldn’t be the first thing we reach for.

The thread running through the whole conversation is that we should be emotionally attached to outcomes, not prescriptions. The goal was never “take 50 milligrams of losartan.”

The goal is that you don’t have a stroke, you don’t end up on dialysis, and you don’t develop heart failure. Between lifestyle levers and inexpensive medications, high blood pressure is largely a solved problem in medical science. The unsolved problem is implementation. It’s hard to fit the solution to an individual patient’s actual life.

That’s the hard part, and it’s also the interesting part.

Lucy is a terrific internist and an even better interviewer, and her newsletter is worth your subscription if you don’t already subscribe.

Give the episode a listen and if it’s been a while since you’ve checked your blood pressure anywhere other than a doctor’s office, consider this your nudge to spend the $40 for a good blood pressure machine.

We discussed an overall approach to how I think about prevention for my patients, covering topics like ApoB, Lp(a), coronary artery calcium, GLP-1s, and how to think about cardiometabolic risk before diabetes is on the chart.

Rather than recap the episode, I want to use this space to say what I was actually trying to argue across two hours, because the bullet-point version of preventive cardiology (”check ApoB, check Lp(a), think about a CAC”) misses a lot of important nuance.

The standard lipid panel is like ChatGPT. It’s often right but sometimes wrong

The standard cholesterol panel often gets things wrong for patients who have prediabetes or diabetes, insulin resistance, or MASLD (previously called fatty liver).

The LDL cholesterol that the regular lipid panel checks underestimates risk in many of these patients.

In someone with central adiposity, elevated triglycerides, low HDL, and rising fasting glucose or insulin, the LDL number can look reassuring while the underlying particle burden is anything but.

The simplest way to describe the discordance is to think of heart disease as a traffic problem: LDL-C tells you how many passengers are on the road. ApoB tells you how many cars are on the road. Cars are what cause traffic.

That’s why I think ApoB is most useful as a concordance check rather than a routine target. Order it to see if it tracks with the LDL. If it’s 10-15% less, then the numbers are concordant and you probably don’t need to keep checking apoB.

If it doesn’t, or if it’s higher than the LDL, then the patient has discordant numbers - and that is most common in exactly the metabolically unhealthy patients who already worry me.

For these folks, ApoB is the number you should be following.

Lp(a) is a good test to use as a tiebreaker for decision making

Roughly 1 in 5 people have an elevated Lp(a). But most people have ever had it measured. It’s almost entirely genetically determined and quite stable across the lifespan, which means a single measurement settles the question for life.

The pushback that some doctors give for checking the test at all is “what are you going to do with that information?”

The honest answer about what to do with an elevated Lp(a) is less satisfying than the case for testing it.

We don’t yet have an outcomes-proven Lp(a)-lowering drug. The drugs in development can drive Lp(a) down 80–100%, but they aren’t FDA approved and I won’t be ready to prescribe them until we see hard endpoints - lowering a lab value is not the same as preventing a stroke.

That data is coming, and I’ll be paying close attention.

In the meantime, Lp(a) is a risk amplifier that changes how aggressively I treat everything else.

A 42-year-old with a borderline LDL and an Lp(a) of 200 is a different patient than the same person with an Lp(a) of 30. I’ll push harder on LDL, lower my ApoB target, talk more urgently about lifestyle, and (in rare cases) suggest aspirin as a prevention tool.

The Lp(a) doesn’t change what tools I have. It changes how aggressively I think about using them.

CAC is a decision tool, not a screening tool

The single most common mistake I see with coronary artery calcium (CAC) scoring is ordering it when the decision is already made.

If you already know you’re going to treat this patient aggressively, a CAC score isn’t going to change much. But it has the potential to add anxiety, downstream testing, and occasionally a diagnostic cascade that ends in a cath lab for a patient who almost certainly did not need to go there.

CAC has a role for the patient who’s on the fence about a blood pressure medication or a cholesterol medication, who wants to know if there’s actually plaque there, who needs something more concrete than a risk estimate to make a decision. In that patient, CAC can be a useful tool.

For a lot of other patients, I think we over-order it.

And - very importantly! - CAC isn’t useful for young people. A CAC of zero in a young patient is a meaningless data point. It changes nothing about risk. Young people are more likely to develop soft plaque, and I’ve seen too many young people with a CAC of zero and alarming amounts of plaque in their arteries to ever trust a reassuring result in someone under age 50.

A related point I tried to make: CAC and coronary CTA (CCTA) are not interchangeable.

CAC is a risk-refinement tool for asymptomatic patients. CCTA is an anatomic test, more appropriate when you’re evaluating symptoms or when you genuinely need to know whether there’s non-calcified plaque. I occasionally send a CCTA in an asymptomatic patient with a very high-risk profile when I think the result will meaningfully change behavior - but that’s a controversial perspective, and certainly not in the guidelines.

Risk calculators miss some of the highest risk patients

So much in medicine is moving to an algorithm for assessment and treatment.

Unfortunately, when it comes to cardiac risk, there’s a really high risk group that gets missed by all of the usual risk assessment tools: the young person at elevated heart attack risk.

Since age is the biggest driver of heart disease risk, every risk calculator treats age as dominant input variable. That means young people who are at genuinely high risk don’t get flagged appropriately.

Risk calculators work across populations, but doctors don’t take care of populations, we take care of people.

Treat the person in front of you, not the somewhat arbitrary number that you get from a risk calculator.

A normal A1c is not a clean bill of metabolic health

A hemoglobin A1c is a marker of average blood sugar over the last couple of months, but it can be misleading about metabolic health.

Think about a patient whose A1c is 5.6, whose LDL is 110, and whose was told everything looks great. But a deep dive into the numbers shows a triglyceride level of 220, HDL is 38, waist is 42 inches, and ALT is creeping up.

That patient already has cardiometabolic disease. Waiting for the A1c to cross 5.7 (to become “prediabetes”) or over 6.5 (to become diabetes) to act is not the right strategy. Metabolic health is a holistic assessment, not just a look at a single metric about average blood sugar.

The triglyceride-to-HDL ratio isn’t a diagnostic test, but it’s a useful tell. So is the trajectory of the numbers. So is the waist circumference. So are liver tests. The job of a preventive doctor isn’t to make a binary diagnosis of metabolic syndrome - it’s to notice the pattern early and give patients the information that may inspire them to take action.

Statins are often first line drugs, but GLP-1s increasingly have a place in a cardiology clinic

For all the noise about advanced biomarkers and new drugs, the most important thing I do for primary prevention is start a statin in the right patient.

The evidence base is overwhelming, the cost is trivial, and most of the “side effects” patients fear are either reversible by switching agents or attributable to other things entirely.

What’s genuinely changing clinical practice is GLP-1 therapy as a cardiometabolic intervention rather than a diabetes drug.

SELECT showed semaglutide reduces MACE in patients with obesity and prior CV disease, and the tirzepatide data is emerging in a similar direction. For the patient with central adiposity, MASLD, sleep apnea, and rising glucose, a GLP-1 may now address multiple risk drivers at once.

Ozempic doesn’t replace the statin. It sits next to it.

The most important take-home point - the patient is a person, not a number in a risk calculator

Most of preventive medicine isn’t a debate about which test to order.

Decision making on treatment is about meeting a patient where they are in terms of risk and in terms of personal preference.

The tools we talked about on the episode - ApoB, Lp(a), CAC, PREVENT, GLP-1s - are just that. Tools. And tools only matter based on how we use them.

Thanks to the Curbsiders team for having me on. Video here:

You name the nutritional conundrum, and I’ve been asked about it.

Here’s a nonexhaustive list of the questions I’ve been asked by patients just in the past week: Can I drink coffee? How much alcohol can I have? What’s the most red meat I should have in a week? Does bacon really cause cancer? Do I really need to eat vegetables?

As I’ve written about before in this newsletter, I used to be fascinated by nutrition research, but the longer I’ve been mentally marinating in these studies, the less I think that there are definitive answers to most nutrition questions.

For today’s newsletter, I wanted to share the latest Beyond Journal Club Podcast where we tackle the thorny topic of the impact of coffee on irregular heartbeats with a look at the CRAVE Trial.

We discussed a lot in this podcast before we get into the question that the CRAVE Trail asked, which is one that patients ask me all the time: does coffee cause irregular heartbeats?

We had the chance to interview Dr. Walter Willet, one of the most prominent nutrition researchers of all time, talked to us about the challenges of studying the big questions in nutrition.

Before we get into the study itself, I wanted to share a few big picture thoughts that we tackle.

Most nutrition research is not worth your time to understand or to think about

One of the reasons that nutrition research is so challenging - and that people are so confused by it - is that many of the questions that people ask in this realm are literally unanswerable with the way that research is done.

When a study is incapable of answering a question like “does red meat cause cancer?” there isn’t much point in diving into the nuances of how the data was collected and what the limitations of the associations found are.

And so you’ll find that consuming the breathless media coverage of a lot of these studies is just a total waste of your time.

It’s also a waste of the time of many of the people who do the research itself. As Adam Cifu, MD has politely requested to these researchers, Please Do Not Do That Research.

Most nutritional research is literally just mindless content that shouldn’t be discussed as though it’s important science.

But every once in a while, we get a study that asks a short term, somewhat answerable question, and that’s why I think that the CRAVE Trial was so interesting.

The question of whether coffee causes irregular heartbeats is a common question that patients have

Some people experience palpitations after drinking coffee and many people have irregular heartbeats.

Answering the question of “does coffee cause arrhythmias” is something that has a ton of practical relevance.

CRAVE tried to ask a question you can actually answer with a controlled experiment. You can’t randomize people to drink coffee for 40 years. But you can randomize them to drink coffee or abstain, day by day, over two weeks, while monitoring their heart rhythms with a continuous ECG patch.

That’s basically what the investigators in this study did. They took a hundred healthy adults, gave them heart monitors, Fitbits, and continuous glucose monitors, and sent them text messages each night telling them whether to drink coffee or avoid it the next day, and the whole thing ran for 14 days.

They collected data on irregular heartbeats, step count, and sleep data.

The headline result was simple: no increase in PACs (premature atrial contractions) on coffee days versus no-coffee days.

The media ran with that: coffee is safe, no irregular heartbeats.

But that headline, while technically true, is incomplete.

This trial actually had results that you could take in a bunch of different ways

The topline result - no irregular heartbeats from coffee - is technically true. But the other outcomes are where things get more interesting.

Coffee days led to more of a different type of irregular heartbeat, PVCs (premature ventricular contractions). There were 154 daily in the coffee group versus 102 on average in the no coffee group.

Coffee days also produced about 1000 more steps per day and 36 fewer minutes of sleep per night. Glucose levels were the same between groups.

So the full picture is more nuanced than the headline suggests. PACs didn’t change. PVCs went up. Physical activity went up. Sleep went down.

Something for everyone.

Do you want to believe coffee is good for you? Well, then it didn’t cause an increase in PACs and it led to 1000 extra steps per day.

Do you want to believe coffee is bad for you? Well, then it caused more PVCs and led to less sleep.

And we don’t know whether these results extrapolate to sicker people or those with preexisting irregular heartbeats, and that’s where DECAF came in.

The DECAF trial asked: what About Patients With Atrial fibrillation (Afib)?

The episode also covers DECAF, which looked at 200 coffee-drinking patients with AFib who were going for cardioversion.

Half were encouraged to drink caffeinated coffee daily for 6 months; the other half abstained, and the coffee group had a lower rate of AFib recurrence - 47% versus 64%.

I’d be cautious about reading too much into DECAF, just like we shouldn’t read too much into CRAVE.

The primary endpoint was clinically detected arrhythmia recurrence, not continuous rhythm monitoring, so lots of short runs of asymptomatic Afib wouldn’t have been captured.

And notably, only 40% of the participants reported that coffee ever triggered their Afib or palptiations in the first place.

But taken together, CRAVE and DECAF are probably the best data we’re going to get on this question, and the consistent message across both is that coffee is probably not much of a cardiac villain.

What I Tell My Patients

My clinical take, which I’ve had in some version of this conversation with a patient basically every week of my life, is simple.

If you like coffee, drink it.

I’m not going to tell you to start it as a therapy, and if it causes palpitations you find bothersome, then you should not drink something that makes you feel unwell.

But there’s no compelling evidence that coffee is harmful for your heart. And so if you like coffee, and you enjoy drinking it, then you can feel free to do so.

I hope you enjoy the podcast, and the next time someone sends you a study about coffee, olive oil, or red wine, hopefully you now have some additional tools to evaluate whether it’s worth your time to consume an influencer’s 45 second take about “what it means for you.”

The study we covered is a trial called QWINT-1, which tested a once-weekly basal insulin for type 2 diabetes.

Most people who have diabetes are terrified that they’re going to need to start injections.

And most doctors who have had the “we need to start insulin” conversation with a patient know the hard part usually isn’t the pharmacology.

It’s the burden that insulin creates for a patient: daily injections, titration, fingersticks/CGM interpretation, fear of hypoglycemia, and the general sense that starting insulin is a line you cross and don’t come back from.

The promise of QWINT-1 is that maybe insulin therapy can be easier - fewer needles, less adjustment - with a really long acting injectable version.

And the top-line answer is basically “yes,” insulin can be more convenient: super long acting insulin efsitora was non-inferior to long acting insulin glargine for A1c lowering at one year.

It should be a bigger deal than this really is.

But the real world implication is way less groundbreaking - QWINT-1 studied a group of patients that would almost never be placed on insulin because they hadn’t come close to maxing out on oral treatment options.

In this episode, we talk about the process of insulin resistance and how diabetes develops, when it makes sense to start insulin, and how to think about long acting insulin in a world where GLP1s and SGLT2s have changed the landscape of diabetes care.

If you noticed some of the news about President Trump taking aspirin and wondered if you should be on it (or if your patients should be on it), then this podcast should help to clarify the evidence base. not alone.

What We Covered:

The 2018 trifecta (ASCEND, ARRIVE, ASPREE) that changed guidelines on aspirin use.

Spoiler: the absolute benefit is vanishingly small, roughly matched by bleeding risk. And for patients over 70, the risk is almost certainly greater than the benefit.

We also cover why traditional risk assessment miss the patients who might actually benefit from aspirin - those with strong family history plus a difficult-to-modify risk factor like high Lp(a), active smoking, inflammatory diseases, or very poorly controlled diabetes/metabolic syndrome.

We also briefly address a topic I’ve written about a bunch before in this newsletter - the triglyceride/HDL ratio as an underutilized marker of insulin resistance that doesn’t show up in any risk calculator.

We get into how aspirin works in heart disease (blocking platelet aggregation at plaque rupture) versus what it doesn’t do (prevent atherosclerosis progression).

If you’re not controlling lipids and blood pressure, taking aspirin is majoring in the minor.

Clinical Pearls:

There’s no benefit in 325mg over 81mg. ADAPTABLE showed no benefit over 81mg for secondary prevention. This makes biological sense since aspirin irreversibly blocks platelet function, and 81mg daily adequately eventually inhibits all circulating platelets. We only load with high-dose aspirin during acute MI because patients need immediate platelet inhibition. The higher bleeding risk at 325mg likely comes from aspirin’s second mechanism: direct disruption of gastric mucosa, not enhanced platelet inhibition.

If you talk to doctors who are experts in platelet biology, they'll tell you that a tiny subset of people may benefit from 81mg of aspirin twice a day instead of daily, because they are just producing platelets so quickly that the more frequent dosing is helpful for chronic antithrombotic impact.

Who might actually benefit from aspirin for primary prevention? Young patients with limited modifiable risk factors but strong family history - think high Lp(a), active smoking, or active inflammatory diseases like lupus. These are patients where you don’t have many other levers to pull, but are also at increase thrombotic risk and lower bleeding risk (mainly due to age).

Could clopidogrel be safer? Interesting question. Aspirin increases bleeding through two mechanisms: platelet inhibition and gastric mucosal disruption. Clopidogrel only affects platelets, which raises the possibility it might have a better safety profile for primary prevention, alhough we lack the trial data to support this.

Don’t throw the baby out with the bathwater. Weak evidence for primary prevention doesn’t negate strong evidence for secondary prevention. Patients with prior MI, stroke, or coronary stents should almost certainly remain on aspirin.

Focus on what actually matters. Treating hypertension and hyperlipidemia is significantly more important than adding a blood thinner for most patients. When a topic in medicine feels confusing, it usually means the magnitude of benefit or harm is small, and we’re better off focusing our energy elsewhere.

If you want to read what I wrote about aspirin a few years ago after the new USPSTF guidelines came out, you can see it here

:

The Bottom Line:

For most patients, aspirin for primary prevention offers minimal benefit with significant bleeding risk.

When I see an elevated fasting triglyceride level, even if everything else looks normal, I do a few things.

First, I confirm it was actually fasting (non-fasting TGs are almost always elevated). But if it’s high and it was fasting, that means we should take action.

Doing nothing is the wrong thing, but action doesn’t necessarily need to come in the form of a prescription.

Here’s what I tell patients: high triglyceride levels mean that you’re at elevated cardiovascular risk. And some of this risk is in your control. I want you to feel empowered by that.

Action can come in multiple forms:

* You can improve your diet - eliminate calories that you’re drinking (juice, sweetened teas, soda, and even alcohol), cut down significantly on refined carbs, added sugars. Even just reducing overall calories will have an impact

* You can get more active - increase overall physical activity (even a small thing like a 15 minutes walk a couple of times a day makes a difference) or you can focus on building muscle with resistance training

* If your sleep isn’t great, work on sleep hygiene. If you snore, get checked for sleep apnea.

I’ve seen patients drop their triglycerides more than 50% with lifestyle alone. And you can see these changes in a few weeks after they’re implemented.

Medications fall into two buckets:

* Lipid-targeted therapies: statins, PCSK9 inhibitors, icosapent ethyl (trade name Vascepa, which I’ve written about here), fibrates, ezetimibe

* Weight-loss medications: GLP-1s like Zepbound and Wegovy

Both approaches work, often in combination.

The podcast goes deep on when to worry about pancreatitis (inflammation of the pancreas that can occur when triglycerides are over 500-600 mg/dL), specifics about how to counsel on diet and exercise, which medications move the needle most (see the graphic below), and when to think about genetic causes.

The key point - triglycerides are a marker of increased risk, they aren’t the target of treatment

Triglycerides are a marker of increased cardiovascular risk, but they’re probably not a great target of medical treatment.

I’ve written about this before.

Trials like PROMINENT that specifically lowered triglycerides with fibrates showed no cardiovascular benefit.

This is a theme I keep coming back to: don’t fall in love with the biology. Just because something makes mechanistic sense doesn’t mean targeting it clinically works.

So what should we do?

Triglycerides are a call to action to lower cardiovascular risk.

This is partly because high triglycerides often mean we need more apoB particles to carry cholesterol around our bodies (the cholesterol that the particles need to carry gets crowded out by triglycerides and we need to make more particles) and partly because high triglycerides are often a marker of metabolic syndrome (especially when they travel with low HDL).

High triglycerides mean we should do something to lower heart disease risk, not just to lower triglycerides.

You can see the transcript and show notes here.

This is worth your time if you have high triglyceride levels, or if you see patients with metabolic syndrome, diabetes, or elevated TGs you've been meaning to address.

It’s a common problem to see in the hospital - someone gets admitted with an infection, and when we culture their blood, we see bacteria in there.

This can be a scary situation, and the impulse of many doctors is that longer is better when it comes to antibiotics in these cases.

For a while, evidence has been accumulating that a shorter course of antibiotics might be just as good as a longer one, albeit in small trials rather than big ones. But the direction of the evidence has been pretty consistent.

The BALANCE trial, recently published in NEJM is the largest study ever done on antibiotic duration for bacteremia.

I’ll skip to the top line result quickly - shorter courses of antibiotics seem just fine in most cases.

But if you’re just asking the question “is a 7 day course of antibiotics in bacteremia good enough, then you’re asking the wrong question.

Or, at least, you’re asking an incomplete question.

The most notable thing that I heard when preparing for this podcast came from one of the ID doctors we interviewed, who told us:

“I feel somewhat uncomfortable discussing bacteremia as a generalizable thing divorced from nuances.”

That’s the most important point in the podcast.

Asking “how long should antibiotics last?” without specifying the host, organism, and source means that you’re asking an unanswerable question if you want to take good care of the patient.

That clinical context really matters.

A Better Bacteremia Framework

Like many problems in medicine, antibiotic duration for bacteremia is a fractal problem.

Each question you need to answer has a whole bunch of downstream questions that also need to be answered, otherwise you’re making a decision without considering the entire picture.

Before you can meaningfully talk about antibiotic duration for bacteremia, you need to think about three things:

The Host: Are they hemodynamically stable? Do they have hardware? Are they immunocompromised? How quickly did they improve?

The Organism: Staph aureus behaves completely differently than E. coli. Strep viridans carries different endocarditis risk than Strep pyogenes.

The Source: Was source control obtained? Is there osteomyelitis? What about endocarfitis? Cellulitis vs. abscess makes a difference too.

The episode walks through each of these in detail, with concrete clinical examples of when you’d lean toward 7 days versus when you’d be thinking a longer duration makes more sense despite what the trial says.

What This Trial Actually Shows

BALANCE randomized patients with various types of bacteremia to either 7 or 14 days of antibiotics. Most had uncomplicated gram-negative bacteremia, but about 20% had gram-positive organisms.

The 7-day arm had a mortality rate of 14.5% versus 16.1% in the 14-day arm. Non-inferior.

Secondary outcomes like ICU admission, relapse, C. diff, length of stay were all similar between groups.

The median treatment in the “7-day” arm was about 8 days because clinicians were making individualized decisions based on patient factors, organism characteristics, and clinical response, which is exactly what we should be doing.

Paradigm Reinforcing, Not Paradigm Changing

This is one of those trials that doesn’t upend clinical practice but rather confirms that the evidence has been nudging us in the right direction.

A trial doesn’t have to have a surprising result to be important. It’s really reassuring to know that those smaller studies haven’t been misleading us.

We’ve already had trials showing shorter courses work for ventilator associated pneumonia, pyelonephritis, community acquired pneumonia, cellulitis, intra-abdominal infections.

BALANCE adds bacteremia to that list with the largest patient cohort we’ll probably ever see.

What this means to me is that the default should now be 7 days of antibiotics for uncomplicated gram-negative bacteremia in immunocompetent patients.

But you still need to individualize treatment. It’s important not to be too concrete about a trial result like this. Longer courses are still warranted for:

* Staph aureus bacteremia

* Endocarditis

* Osteomyelitis

* Persistent fever despite appropriate antibiotics

* Immunocompromised hosts

* Inadequate source control

Why This Trial Matters Even If It’s “Expected”

One of my co-hosts asked: if we already suspected these results, was this trial worth the resources?

My answer: absolutely yes.

There’s a huge difference between “we think this is probably true” and “we feel much more confident that this is true based on robust data.”

Small trials can point in one direction, then a large trial in the modern era comes along and completely reverses what we thought we knew. (Think MINT trial on transfusion thresholds, which we covered in a previous episode.)

BALANCE reassures us that the direction of the evidence has been correct and we don’t need to backtrack.

Listen If You Want to Actually Understand This Trial

The episode is worth your time because we go deep on the nuances that get lost when people just quote the headline.

We discuss:

* Which organisms weren’t included (HACEK, TB, Listeria, Capnocytophaga, Staph aureus)

* How to think about gram-positives in this trial

* The difference between trials that change paradigms versus trials that reinforce them

* Why antibiotic stewardship needs these kinds of large confirmatory studies

Listen wherever you get your podcasts, or read the full transcript here.

I’d love to hear your thoughts, especially if you’ve been in situations where the “right” antibiotic duration wasn’t obvious and you had to BALANCE (sorry, couldn’t resist) competing considerations.

Thank you for reading! Please share with friends and family and encourage them to subscribe!

Disclaimer: This is my opinion, not medical advice. Reading this newsletter does not constitute the formation of a doctor-patient relationship and is no substitute for the opinion of your doctor.

It’s not that much of an exaggeration - more than 1 out of 4 patients are dead within a year after a hip fracture related to osteoporosis.

Most people have heard of osteoporosis, and a lot of patients (even young ones) are vaguely aware of the importance of bone density as we age.

Here’s the biggest thing that shocked me while researching this episode: 73% of fractures occur in people who don’t have osteoporosis.

Read that again. Three out of four people who break their hip or spine would have had normal bone density scans or mild osteopenia before their fracture, and they wouldn’t have been candidates for treatment under current guidelines.

That’s a massive diagnostic and therapeutic gap.

In the latest episode of Beyond Journal Club podcast, we cover a new trial on osteoporosis treatment, which may totally change the paradigm of preventing fractures.

This new trial, published earlier this year in the New England Journal of Medicine, asks a provocative question: What if we just treated postmenopausal women with bisphosphonates to prevent fractures rather than waiting until they seemed high risk or developed frank osteoporosis?

This study looked at a really low risk group of patients - no evidence of osteoporosis on bone density scans. And the medical intervention they used is pretty minimalist - 1 or 2 doses of zolendronate over a period of 10 years.

You read that correctly. They followed patients for 10 years, and the intervention group either received one dose at the onset of the trial or one dose at the onset of the trial and another at 5 years.

The results are compelling. In healthy women in their 50s, that a single dose of IV zoledronate (or two doses over 10 years) reduced vertebral fractures by about 30-40%. The number needed to treat was 20 to prevent one vertebral fracture.

And here’s another wild piece of information from the trial: in the placebo group, one in three women had a fracture over 10 years. These were healthy, active women who responded to letters in the mail asking them to join a study. Not exactly a high-risk population.

The episode covers a lot of ground:

* Why our current diagnostic approach to osteoporosis misses most people who will fracture

* The full toolkit of fracture prevention medications and their trade-offs

* How to think about the rare but scary side effects (atypical femoral fractures, jaw osteonecrosis) in context - hint: they’re not very common and way overblown by most popular coverage

* Whether this trial pushes treatment for osteoporosis into the primary prevention space

* The public health versus individual patient perspective on treating millions of asymptomatic women

One major idea that came up in our discussion that I’m going to take with me as I think about this for patients: maybe we should start calling fragility fractures “bone attacks.”

Bone density problems are similar to heart disease in some ways. Just like atherosclerosis builds silently until you have a catastrophic heart attack, osteoporosis progresses until boom - you have a devastating fracture that changes the trajectory of your life.

If we framed it that way, would we take prevention more seriously?

This is an episode where I genuinely changed my thinking while preparing it. The question isn’t “does this person have osteoporosis?” It’s “is this person at risk for a bone attack?”

Listen wherever you get your podcasts, or read the full transcript here.

Would love to hear what you think, especially if you work in this space or have personal experience with osteoporosis treatment decisions.

The trial asked a simple but provocative question: Should every patient get a beta blocker after a heart attack, even if they’re doing fine otherwise?

REDUCE-AMI makes us think about one of the most deeply entrenched practices in cardiology.

Below you can find some of my additional thoughts - some of which made it into the podcast, and some of which didn’t. I hope you enjoy the podcast.

Why Beta Blockers Were Always the Default

For decades, prescribing beta blockers after a heart attack has been clinical canon.

The practice is so entrenched that it's a quality metric hospitals are evaluated on. Miss too many post-MI beta blocker prescriptions, and you risk financial penalties.

But here’s the thing: most of the data supporting this practice is from the 1980s and early 1990s. The world of cardiology has changed dramatically since then. Back then, we didn’t have:

* Stents and modern revascularization

* Potent dual antiplatelet agents like clopidogrel, prasugrel, or ticagrelor

* Widespread statin use

* Ezetimibe, PCSK9 inhibitors, or bempedoic acid

We’re playing a different game now. And that means the rules might need to change.

What the REDUCE-AMI Trial Found

REDUCE-AMI is the first large randomized trial to directly ask:

Do beta blockers help patients with preserved ejection fraction after a heart attack?

The short answer: no.

In over 5,000 patients, beta blockers didn’t reduce the risk of death, hospital readmission, or even arrhythmias like atrial fibrillation. These were well-treated patients, on modern therapies, with normal post heart attack cardiac function.

It's the kind of result that makes you take a hard look at what you thought you knew - that old adage they tell you in medical school, that in 10 years half of what you learn will be wrong, we just don’t know which half.

More data is coming, and we can expect other trials over the next year that will influence how firmly we believe the REDUCE-AMI results to be true. But unless they show something wildly different, REDUCE-AMI may mark the beginning of the end for routine beta blockers in this population.

Three Big Lessons from REDUCE-AMI

1. Every Practice in Medicine Needs an Expiration Date

Medical evidence ages. Treatments that made sense in 1985 may not make sense today because so much about the rest of our care changes. That’s not a knock on the past. It’s a sign that each decision needs to be continually reevaluated.

But when our practices don’t evolve with the data, we risk undertreatment, overtreatment, side effects, and missed opportunities to simplify care.

We need to keep asking: Does this still make sense, in the context of how we treat patients today?

2. Sicker Patients Are Sicker

This study enrolled a relatively healthy group: post-MI patients who maintained good ejection fraction. That’s a big deal.

Much of medicine comes down to identifying who is the vulnerable patient that needs us to intervene.

There’s a reason we tell interns that we want you to be able to identify sick versus not sick - deciding which patients are sicker helps us make a lot of treatment decisions on the margins for people who might not fully fall into Table 1 of a clinical trial.

The converse is true too - healthier patients are healthier.

One of the hardest skills in clinical medicine is knowing when not to treat. And REDUCE-AMI is a reminder that if someone is doing well, less may be more.

3. Goodhart’s Law Strikes Again

Goodhart’s Law (borrowed from economics) says:

“When a measure becomes a target, it ceases to be a good measure.”

In healthcare, this shows up all the time. We turn surrogate markers - like LDL levels, blood pressure, or yes, even beta blocker prescription rates - into metrics of quality.

But metrics shape behavior. When beta blockers become a benchmark for good care, doctors are incentivized to prescribe them on the margins and every edge case is more likely to end up with a beta blocker prescription if it changes your quality metrics.

And once you create incentives around a metric, it becomes incredibly hard to walk it back, even if the evidence evolves.

If we don’t update our quality metrics with the same urgency that we update our clinical practice, we’re just going to be algorithm monkeys left chasing numbers that don’t matter.

Why This Matters for the Future of Healthcare

As AI becomes more integrated into care delivery and decision-making, the metrics we choose will matter more than ever.

The trend of metricizing what we do in health care isn’t going away. It’s accelerating. And because AI moves us towards optimization, as these tools become integrated into medicine, expect the issue to get more entrenched rather than less.

And so if our targets are outdated, misaligned, or based on shaky evidence, we run the risk that the further metric-ization of health care will make things worse for our patients.

So when you look at something like REDUCE-AMI, it’s not just a trial about beta blockers. It’s a canary in the coal mine that suggests how flexible our systems are is going to matter a lot.

Whether we can evolve when the data tells us to - and whether we use the right data to guide us - will determine if things evolve for the better.

🎧 Listen to the full episode of Beyond Journal Club wherever you get your podcasts. We dig deep into the trial design, the history of beta blockers post-MI, and what this means for real-world patient care.

I’d love to hear your thoughts—especially if this changes how you think about treatment after heart attacks.

At first glance, the topic might seem almost...boring. HIV has been around for decades. It's still a chronic disease. There’s still no cure, no vaccine — same old story, right?

But as I dug into these trials, I was floored by the bigger picture: the science here is nothing short of revolutionary.

In the span of a generation, we’ve gone from discovering HIV to having the tools to end the HIV epidemic today.

A diagnosis of HIV used to be a death sentence. Now it’s a chronic, manageable condition.

And for people at risk, prevention is easier than ever: a daily pill, a monthly injection, or an injection every six months — all incredibly effective, well tolerated, and (hopefully) widely available.

The hard truth? The obstacles to ending HIV are no longer scientific. We don’t need new drugs or vaccines. We need policy, access, and implementation.

There are 30,000 new HIV infections every year in the United States. That number should be zero.

We made this episode because everyone — policymakers, doctors, and the public — needs to understand what’s possible when we match scientific breakthroughs with real-world action.

The obstacles to ending HIV are no longer scientific. We don’t need an HIV vaccine, we don’t need new medications, we just need to help people at risk get access to the type of preventive medications that work for them.

If you aren’t familiar with this story, I think you’ll find this episode not just illuminating — but inspiring.

If you’ve never had your Lp(a) checked, I think that you should.

The importance of Lp(a) in the cardiac risk assessment is something that I’ve written about and podcasted about many times before.

So for today’s newsletter, I’m sharing a podcast that I was interviewed on that gets into the weeds about Lp(a).

Here’s an overview of what I think is important to know about this test:

* Lp(a) is a blood test that should be done once in everyone’s life. You don’t need to keep checking it

* High Lp(a) increases the risk of heart attack, stroke, and calcific aortic valve stenosis

* Risk increases as levels rise - it’s not positive or negative, Lp(a) risk exists on a spectrum

* The direct treatments for Lp(a) aren’t ready for primetime - they are still in clinical trials - so we don’t have a specific way of treating this issue

* Even though there are no Lp(a)-specific treatments, I still think that checking levels makes sense because it’s actionable in other ways

* High Lp(a) means that we should be treating other aspects of your heart disease more aggressively - if you have high Lp(a) and we’re on the fence about lowering your LDL-C, it pushes me towards treatment. If we’re debating the importance of quitting smoking, it makes me push the conversation a little bit harder.

* In that way, I think of high Lp(a) as a tiebreaker for deciding on treatment in a 50/50 clinical situation

* High Lp(a) also pushes me to do imaging tests like calcium scores or coronary CTAs earlier on so that we can be more precise in our application of other preventive medications

* Lp(a) is mainly governed by genetics - it’s not high because you eat too many seed oils

* Lp(a) increases cardiac risk because it accelerates cardiovascular disease, increases oxidation, and increases blood clotting

* There is some data that aspirin may make sense for preventing a first heart attack or stroke in patients with elevated Lp(a), but the quality of the data isn’t high enough to make a blanket recommendation.

Knowledge is power and understanding your Lp(a) will help you frame your overall cardiac risk.

I hope you enjoy the podcast - and there’s a transcript at this link if you prefer reading to listening.

And everything that you see sounds totally ominous.

Almost all of the content takes it for granted that *obviously* microplastics cause disease, and so *of course* we should be strategizing ways to remove them from our lives.

But is the evidence really all that persuasive?

This is a podcast on the APACHE study - a paper published in the New England Journal of Medicine on the link between microplastics seen in cardiovascular plaque and subsequent development of bad outcomes.

For the podcast, we had the chance to interview Dr. Phil Landrigan, who wrote the NEJM editorial on the paper, as well as Dr. Jane Leopold, the Deputy Editor of NEJM responsible for shepherding this paper to publication.

The podcast is the latest installment of the Beyond Journal Club series that I’ve been working on with Core IM and the NEJM Group.

This is probably the most important study to date on the question of microplastics and human disease

This study looked at people who had surgery to remove a blockage in their carotid arteries and found that more than half of them had actual microplastics in their artery plaques.

The people who had plastic in their artery plaques had more than 4 times the risk of bad cardiac outcomes compared to the group that didn’t have any plastics in there.

It’s the first study that ever demonstrated a link between plaque visually seen in tissue and subsequent risk of disease - this is a really important paper and a big deal in terms of building the science.

And while it sounds totally damning - like this should be the smoking gun linking microplastic exposure to disease in humans - I don’t think that’s the right takeaway.

There are quite a few reasons that I’m skeptical this study proves microplastics cause human disease:

* We don’t know what makes people who have microplastics in their carotid plaques different than people who don’t have microplastics in those plaques. It’s just as plausible to me that sicker people are more likely to incorporate microplastic into their plaque and so the presence of microplastics is actually just a marker of someone’s health status.

* We have no idea about the exposure differences - do people without microplastics all have glass food storage containers and all cotton clothes while the people with microplastics microwave plastic tupperware and drink from disposable water bottles?

* There’s no dose response relationship seen in the plaques - higher levels of microplastic should lead to more inflammation if they are causative, but in this paper that relationship didn’t exist.

* We don’t have any evidence of a temporal relationship between plastic exposure and subsequent risk of disease. That relationship - exposure preceding disease - is part of the Bradford Hill Criteria that help us understand how likely a correlation like this is to be causative.

I don’t think that this study is a “call to action” for doctors

Many people looked at this paper as a “call to action” that “demands urgent attention.”

I’ve had colleagues describe talking about trying to reduce plastic exposure with our patients as an “easy win.”

I disagree with that comment.

While the evidence raises questions, it’s not accurate to say that we know plastic causes human disease.

I’m not there yet, and I don’t think that a microplastic discussion has any place in a doctor’s appointment unless a patient brings it up.

And even if a patient wants to talk about that question, I try to redirect the conversation.

We have a much higher level of confidence that exercise is good for you than that plastic is bad for you.

We know that treating high blood pressure lowers your heart disease risk and we don’t know whether avoiding plastic tupperware has any impact.

There are just so many places where the quality of the evidence gives us clear guidance about what to do and the microplastic question is certainly not one of them.

We need to be disciplined in the way that we interpret the science here. The chain of causality hasn’t been built, and so maybe microplastics cause human disease, but maybe they simply don’t.

But even if you are persuaded that microplastics cause human disease, it’s not clear what you should do about it

The biggest place where I think we are left with questions is what to do about this.

Removing plastic from your life is expensive and inconvenient - and, more importantly, we don’t know how much that impacts your total plastic exposure.

Plastic is everywhere on Earth, from the deepest parts of the ocean to the highest mountains of the Himalayas.

We get plastic into our bodies from ingestion, inhalation, and skin exposure.

None of the evidence points to a clear direction on where the bulk of that exposure comes from.

On top of that, it’s not clear that all plastics are equal when it comes to biologic impact.

There’s an argument to be made that if we, as physicians, are thinking about changing our own behavior that we owe it to our patients to share that with them.

And I understand that point, even if I disagree with it.

We should be rigorous about the way we interpret evidence for our patients - and unless a discussion about microplastics is filled with caveats - I strongly believe that the opportunity cost of discussing this stuff during a clinic appointment is too high.

Remember, time is limited, and what you talk about is what you prioritize.

I’m going to save my time with my patients to discuss the things that I am more confident about in two ways:

* I feel more certain that there’s a direct link between the risk factor and cardiovascular disease

* I feel a higher level of confidence in the things that we can do to mitigate that risk

So until that evidence gets stronger, I’m going to stay focused on discussing heart disease prevention in a way that reinforces my level of confidence in the level of evidence.

There’s a meme in medicine called “things we do for no reason” to describe this tendency for a practice to become ingrained into the culture of what we do in the hospital even when we haven’t generated great evidence to support that practice.

And it might surprise many non-medical readers to know that when doctors are trained and become part of the guild of medicine, we frequently carry on these practices without ever asking about the evidence to support them.

We recently released another episode in our Beyond Journal Club Podcast looking at a topic that will feel like inside baseball to anyone who isn’t involved in medicine: the use of acid suppressing medications in the intensive care unit for intubated patients.

This episode looks at the REVISE trial in detail - it’s a study evaluating a practice that almost everyone who has ever worked in the hospital is intimately familiar with.

Patients who are critically ill are at risk for a lot of bad things to happen while they’re hospitalized, and these patients are often really complicated.

As a consequence, most ICUs operate with an ICU checklist of automatic things that are discussed everyday - things like preventing blood clots, mobilization, feeding, and preventing stomach ulcers.

This podcast looks at the evidence behind preventing ulcers in critically ill patients.

I started out not being really sure if the practice of trying to prevent ulcers in the ICU was one of those things we do for no reason.

But thankfully, after I dug into the evidence, it seems to me that this is a pretty reasonable practice - giving medication to reduce ulcer formation improves outcomes that patients care about.

Working on the podcast also inspired me to try to ask the question “how good is the evidence for this thing that I do all the time” more frequently.

It’s often some of the most common - and the most mundane - questions in medicine that we don’t have great answers to.

We owe it to our patients to be consistently demanding that the evidence base supporting our decision making is strong and we should be comfortable thinking that our clinical practices should have an expiration date, especially as medicine and technology change the context in which we perform them.

In the podcast, you’ll learn about:

* How common are stress ulcers in the GI tracts of critically ill patients?

* What is “clinically important bleeding?”

* What are risk factors for stress ulcer bleeding in ICU patients?

* The evidence behind the use of medications like PPIs and H2 blockers for preventing stress ulcers

* A deep dive into the REVISE trial

* A detailed discussion of whether these results apply to all ICU patients or we should be excluding the sickest ones

I hope you enjoy!

It’s a fascinating piece of data to add to what a lot of us are seeing in clinical practice: this class of drugs works in the brain to change behavior - the reason they cause weight loss is because of the way that they tell our brains that we’re full.

A paper like this is a reminder of something I’ve written about a few times before: Ozempic and the other incretin mimetics are changing the world in a lot of important ways that we’re still beginning to see and understand.

For today’s newsletter, I am re-releasing a podcast about the story behind the most important class of drugs that’s ever been developed.

This is part of a project that I’m working on called “Beyond Journal Club,” a podcast collaboration between Core IM Podcast and the NEJM Group.

We were just picked up for our second season of the podcast, which is a really exciting collaboration that I think has quite a bit of value in the medical education space.

The purpose of the podcast is to take a common medical learning tool, “Journal Club,” and elevate it, going beyond the usual deep dive of methods, figures, and tables.

We aim to tell the story behind a landmark clinical trial to help the results make sense in context.

The podcast really echoes two themes of the newsletter:

* Context matters when you interpret any research. Without context, the information is just as likely to mislead as to be helpful

* Not everything that’s published is important. By understanding results and the context in which they exist, you can understand

I hope you enjoy our second episode on the SURMOUNT-1 trial, which is the clinical trial that put tirzepatide (which you may know as Zepbound or Mounjaro) on the map as a weight loss drug.

This is the story behind the clinical trial.

In the podcast, you’ll learn about:

* A framework for thinking about obesity as an environmental disease, and why this framework helps us understand the rising numbers on obesity and metabolic syndrome

* The history of weight loss drugs before the current era - a story of unintended side effects, ineffectiveness, and not making a dent in the problem

* How the Gila Monster - a venomous lizard native to the American Southwest and Mexico - inspired the development of Ozempic and Mounjaro

* Why the scandal about the cardiovascular toxicity caused by the diabetes drug Avandia changed the nature of diabetes trials and opened the door for the weight loss drug revolution

* The outcomes from the early studies on drugs like Ozempic and Mounjaro, including a head to head comparison

* A deep dive into the SURMOUNT-1 trial, which showed a pretty astounding impact of tirzepatide on weight loss

I hope you enjoy!

Some of our graphics from the show notes below, and the show notes themselves can be found here.

The mechanisms that these drugs use is fascinating - they work all over our bodies, but the weight loss impact probably happens in the brain, where these medications reduce appetite:

The history of landmark trials of these medications on both diabetes and weight loss, as well as a peak into the future:

And that adage is particularly applicable to blood transfusions, where the question of how much blood to transfuse in a patient is a really tough one to answer.

Blood transfusions have undergone a pretty astounding paradigm shift over the last generation - most doctors practicing today would say that when it comes to blood transfusions, less is almost always more.

While we used to transfuse patients pretty liberally, a series of trials over the last 25 years have established that a liberal transfusion strategy isn’t better for most really sick patients.

I wrote about this question a few months ago, wondering “Why was I wrong about blood transfusions” when I dove into the MINT trial in detail.

For the latest episode of our Beyond Journal Club podcast series, we tackled this trial in audio form, diving a bit more deeply into the data and biology.

We interviewed two of the lead investigators from the trial, Jeffrey Carson and Sunil Rao, as part of the prep, and some of their clips are in the podcast.

In the podcast, we cover:

* Why blood transfusions make sense, but also why they might be risky

* What makes a blood transfusion decision during a heart attack different than a decision in other clinical scenarios

* The history of blood transfusion trials

* The MINT trial in detail with a look at the protocol and results in depth

* What we should all take away from this trial.

You can review the show notes in detail here. Below is our graphic on the landmark blood transfusion clinical trials.

I hope you enjoy!

“A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die and a new generation grows up that is familiar with it”

I was reminded of this concept while working on a new Beyond Journal Club podcast about a clinical trial on asthma.

Asthma is an incredibly common disease - we all know someone who has asthma and uses an inhaler from time to time.

One of the things that interesting about asthma is that even though it’s incredibly common and can be a source of major symptoms, it doesn’t kill people very often.

And since most doctors have seen asthma so commonly and we feel comfortable that the majority of patients are going to do ok, we often prescribe the wrong treatment: a rescue inhaler with albuterol.

This is a podcast on the MANDALA trial, which is latest in a long line of clinical trials telling us that an albuterol rescue inhaler by itself is the wrong asthma treatment for most patients with asthma.

Albuterol alone as a rescue treatment just treats symptoms in the moment but doesn’t alter the trajectory of the inflammation in asthma

There’s evidence from quite a few clinical trials that a combination inhaler that combines albuterol with a drug that reduces inflammation is better than just giving people albuterol alone.

In other words: lots of doctors who have been treating asthma for years (myself included) have been doing it wrong!

In this podcast, we go through some of the history of asthma treatment and discuss how our understanding of the disease has changed over time.

We had the chance to interview Dr. Jeffrey Drazen, former editor-in-chief of the New England Journal of Medicine, who is one of the world experts on asthma management.

Dr. Drazen makes a couple of appearances in the podcast to discuss the evolving paradigm of asthma management.

You can review the show notes and even read a transcript of the podcast here.

Here’s a graphic of some of the key trials that have established practice in this area:

I hope you enjoy!