A weekly health led newsletter grounded in evidence-based medicine along with prospective randomized controlled trials (RCTs) by medical specialists. Our goal is to help you make sense of complex scientific information and turn it into clear, evidenced based practices you can use to make better decisions about your health and wellness.

Two people start the same drug on a Monday.

One loses twenty kilos. The other loses four, decides it’s not working, and keeps taking it mostly because the prescription is already filled.

Three years later, their hearts are in roughly the same shape.

That is not a thought experiment. It is what the largest trial ever run on these drugs found, and cardiologists are still arguing about why.

THE 60 SECOND VERSION

WHERE THIS STARTS

In November 2023, a trial called SELECT put 17,604 adults on semaglutide or a placebo. All of them had heart disease and extra weight. None of these participants had diabetes.

Heart attacks, strokes and cardiovascular deaths landed at 6.5% on the drug versus 8% on placebo. A 20% relative reduction, and the first real evidence of GLP-1 benefits beyond weight loss in a trial built to test hard outcomes.

Everyone read that as a weight story. Lose weight, and ultimately spare the heart.

Average weight loss in SELECT was 9.4%. Bariatric surgery actually doubles that. Lifestyle trials have hit similar numbers for decades and never moved cardiovascular events.

The arithmetic did not fit, and the investigators said so themselves.

THE TEST

A mediation analysis asks a simple question using complicated maths. If a drug works by shrinking people’s BMI, then the people who shrank most should have had the best outcomes.

In October 2025 the SELECT team published exactly that analysis.

Among people on semaglutide, the weight lost by week 20 had no linear relationship to their heart risk over the years that followed. Someone modestly overweight got the same relative protection as someone with severe obesity.

Waist circumference did track with outcomes, and it accounted for about a third of the effect.

THE BIOLOGY

Ask most people what causes a heart attack and you get the typical “plumbing answer”. Cholesterol builds up, the pipes narrow, eventually leading to minimal blood flow.

That picture is wrong in a way that matters here.

Most heart attacks do not happen at the narrowest point in an artery. They happen where a plaque cracks open. A plaque is less like limescale in a pipe and more like a wound in the artery wall that never healed, with immune cells camped inside keeping it irritated.

When the thin cap over that wound tears, blood meets the contents underneath and a clot forms in seconds.

This gives you two ways to reverse engineer to prevent a heart attack.

Cholesterol has owned this conversation for forty years. In 2017 a trial called CANTOS tested the other lever alone. Ten thousand people who had already survived a heart attack, all of them already on statins, given a drug that blocks one inflammatory signal and does nothing whatsoever to cholesterol.

Heart attacks, strokes and cardiovascular deaths fell 15%, although LDL did not budge.

This is the reason why the semaglutide inflammation numbers are interesting rather than incidental.

C-reactive protein is what your liver produces when those immune signals run hot. In SELECT trial, it fell 37.8 percent versus placebo.

By week four, it was already down about 12%. By week eight, about 20%. Body weight at those same clinic visits had moved 2%-3%.

RULING THINGS OUT

If the benefit came from cholesterol or blood pressure, the arithmetic would show it. It does not.

READ THE NUMBER PROPERLY

Neither is disappointing. Both are worth knowing before somebody tells you these drugs cut heart attacks by a fifth.

The kidney trial is the stronger of the two, which is the opposite of how the coverage has run.

ORGAN BY ORGAN

Kidneys filter blood under pressure. Run them at a high pressure for twenty years and eventually the filters scar.

GLP-1 drugs allow the kidney to dump sodium within days of the first injection, which drops the pressure inside those filters long before a single kilo comes off. That is the mechanism behind FLOW.

That mortality result matters because it survived the statistical hierarchy of its own trial. The equivalent number in SELECT did not, which is why you will not see it used as a claim that these drugs extend life.

Heart failure with preserved ejection fraction is a heart that squeezes fine but has gone stiff and cannot fill. Fat around and inside the heart is part of reason for the preserved efficiency.

That is what SUMMIT targeted. In the earlier STEP-HFpEF trial, semaglutide added 21.5 meters to how far people could walk in six minutes.

Sleep apnea is the most mechanical win on the list. Less fat around the throat, less collapse at night, and an FDA indication granted in December 2024.

THE QUESTION KEEP COMING UP

This is the question that people are most interested in.

Lean mass on a body scan is not muscle. It is everything that is neither fat nor bone: muscle, but also organs, connective tissue, and even the water inside your cells.

 Both fluid and glycogen are shed in the first months of eating less and the scan records lean mass loss without a single muscle fiber being lost in the reading.

In the STEP 1 scan substudy, 140 people had imaging at week zero and week 68.

Fat mass fell 19.3 percent. Visceral fat, the dangerous kind wrapped around organs, fell 27.4 percent. Lean mass fell 9.7 percent.

Because fat left more rapidly, lean tissue ended up as a larger share of the body than when they began testing. The Tirzepatide substudy landed at 75 percent fat to 25 percent lean mass, the same split ordinary dieting produces.

In 106 adults followed through a year on semaglutide, grip strength went up 4.5 kg and the proportion meeting the definition of sarcopenic obesity fell from 49 percent to 33 percent.

TURN IT INTO BEHAVIOR

None of that happens automatically. Four bodies including The Obesity Society now say resistance training belongs alongside the medication, not after it.

**The randomized trial testing these points mentioned above is enrolling now, with 232 adults split across training, protein, both and neither. LEAN-PREP should settle it.

READER PULSE

Half the conversation about these drugs happens in whispers, and most of it is about the wrong thing. We would rather hear what is really happening.

This is health journalism, not medical advice. These are prescription medicines with real side effects, and in SELECT 16.6 percent of participants stopped treatment because of them against 8.2 percent on placebo. Decisions about starting, continuing or stopping belong with your own clinician.