Skip to content
GLP Loss
← Research
Evidence

GLP-1s and Statins: Three Studies, Three Directions

Dulaglutide cut atorvastatin exposure 21%. Oral semaglutide raised rosuvastatin exposure 41%, outside the prespecified window. Neither changed a label — and the weight loss itself moves triglycerides by 15% while LDL cholesterol moves 2% to 4%, or the wrong way entirely in people with diabetes.

Owen Castellanos9 min read
Three statins, three directionsExposure ratio with the GLP-1, versus the same statin aloneno-effect windowDulaglutide and atorvastatin, AUC ratio 0.79Ecnoglutide and rosuvastatin, AUC ratio 1.06Oral semaglutide and rosuvastatin, AUC ratio 1.410.801.25None of the four changed a dose instruction on any label.And the weight loss moves triglycerides, not LDL cholesterol.

A statin is the drug a person on a GLP-1 is most likely to already be taking. Roughly nine in ten participants in the cardiovascular outcome trial that made the class famous were on lipid-lowering therapy at randomization.[1] So the question of whether the two belong in the same week is not hypothetical for a small subgroup; it is the default case. What the published record shows is narrower and stranger than either a reassurance or a warning would suggest.

The statin is the drug the stomach holds up longest

The mechanism everyone reaches for is delayed gastric emptying, and the general pattern it produces across this class — a lower, later peak with total exposure broadly intact — is set out in the oral medications article. One modeling study put a statin and a simple analgesic through the same analysis at once, and the two came out nothing alike. With semaglutide co-administered, the absorption rate constant for paracetamol fell 53%, adding a five-minute delay to its disappearance from the absorption compartment. For atorvastatin the absorption rate constant fell 72% and the transit compartment rate fell 91%, producing a 67-minute delay. The authors concluded the effects were not of clinical relevance.[2]

That ratio — 67 minutes against five — is the point. Atorvastatin is a low-solubility, high-permeability drug, and the systematic review that sorted this literature by solubility class found the same signature across all four classes it examined: unaffected or reduced peak concentration, delayed time to peak, and no clinically significant change in area under the curve.[3] The stomach does hold a statin back longer than almost anything else measured. It does not obviously matter, because a statin is titrated against a lipid panel drawn weeks later, not against a plasma concentration drawn that afternoon.

Four studies, four molecules, and a spread nobody reports

Subcutaneous semaglutide at steady state was tested against a single 40 mg atorvastatin dose in healthy subjects. The exposure ratio fell inside the prespecified 0.80 to 1.25 interval and peak concentration was not meaningfully changed.[4] Dulaglutide was tested the same way and the area under the curve was reduced by 21%, with peak concentrations generally lower and time to peak trending later.[5] The oral semaglutide program tested rosuvastatin and found the opposite: an AUC ratio of 1.41 (90% CI 1.24 to 1.60), entirely outside the no-effect interval, which the absorption enhancer alone did not reproduce.[6] A newer analog, ecnoglutide, came in at 1.06 (90% CI 0.94 to 1.20) against the same statin in 28 volunteers.[7]

Down 21%, up 41%, and two studies in the middle. Every one of those papers concluded that no dose adjustment is required, and every one of them was right on its own terms — but a page that summarizes them as “no interaction” is describing a consensus the numbers do not contain. The honest version is that the direction depends on which GLP-1 and which statin, that the largest single change was an increase rather than the decrease the delayed-emptying story predicts, and that the drug which produced it is a tablet whose absorption chemistry has nothing in common with the injections this market sells.

Why a 41% exposure change did not become an instruction

A statin dose is not chosen to hit a blood level. It is chosen to hit an LDL cholesterol target, verified by a test that a person on a statin is already having drawn. An exposure change of this size inside a drug with a wide therapeutic margin shows up, if it shows up at all, as a lipid panel that has moved — which is the thing being measured anyway. That is the structural reason none of the four studies produced a monitoring instruction, and it is a much better reason than “the interaction is small,” because in one case it was not small.

What the weight loss does to the panel, split by column

The more common assumption is that dropping fifteen percent of body weight fixes the lipid panel and makes the statin conversation moot. The semaglutide label’s own tables say otherwise, and they say it differently in different trials. At week 68, the relative difference from placebo in LDL cholesterol was −3.8% in the obesity trial and −7.1% in the trial run alongside intensive lifestyle therapy. In the trial of patients with type 2 diabetes it was +0.4% — LDL rose 0.5% from baseline on drug and 0.1% on placebo.[1]

Triglycerides behaved nothing like that. Across the same three trials the relative differences were −15.8%, −17.0% and −13.9%, and in the cardiovascular outcome trial at week 104 the pattern held: total cholesterol −2.8%, LDL −2.2%, HDL +4.2%, triglycerides −15.6%.[1] A GLP-1 is a triglyceride drug and a statin is an LDL drug, and the subgroup where the LDL effect reverses is the one whose members are most likely to already be on both. The same split turns up in the liver, where the triglyceride and hepatic fat response is the measurable one; that is the subject of a separate page.

The outcome trial was run on top of statins, not instead of them

Among 17,604 participants with cardiovascular disease and no diabetes, 90% were on lipid-lowering therapy at baseline, 86% on a platelet aggregation inhibitor, 74% on an ACE inhibitor or angiotensin receptor blocker, and 70% on a beta blocker.[1] Whatever cardiovascular benefit that trial demonstrated — the arithmetic is set out in the heart benefits article — it was demonstrated as an addition to a statin, in people who kept taking one. Nothing in that trial tested stopping.

The symptom-attribution problem points both ways

Muscle aches, fatigue and general malaise arrive together during a GLP-1 dose escalation, and a statin is the standing suspect for all three. A 2026 meta-analysis of individual participant data from 19 double-blind trials — 123,940 participants, median follow-up 4.5 years — tested every undesirable effect listed in statin product labels. Beyond the already-established muscle and diabetes effects, only four of 66 further outcomes reached significance after controlling the false discovery rate: abnormal liver transaminases (RR 1.41, 95% CI 1.26 to 1.57), other liver function abnormalities (RR 1.26, 1.12 to 1.41), urinary composition alteration (RR 1.18, 1.04 to 1.33) and edema (RR 1.07, 1.02 to 1.12). The authors concluded the blinded data do not support causal relationships for most of the listed effects, including cognitive impairment, depression, sleep disturbance and peripheral neuropathy, and called for the labels to be revised.[8]

Read alongside a drug that reliably produces fatigue and reliably takes lean mass with the fat — the subject of the muscle loss article — that finding cuts against the intuitive move. Stopping the statin because something aches is a decision made against a base rate that randomized evidence says is largely not the statin, while the new drug in the week is the one with the documented tolerability curve.

What the labels actually contain, which is nothing

Neither the semaglutide nor the tirzepatide weight-management label names a statin anywhere. The semaglutide label carries one general instruction: the drug “causes a delay of gastric emptying and thereby has the potential to impact the absorption of concomitantly administered oral medications,” with monitoring advised for medications that “have a narrow therapeutic index or that require clinical monitoring.”[1] The tirzepatide label instructs that caution be exercised with oral medications and names warfarin, not a statin, as its narrow-therapeutic-index example.[9] The atorvastatin data sit in a figure in the diabetes label’s clinical pharmacology section, where the drug is listed among those assessed after a single dose.[10]

A statin is neither narrow-index nor concentration-monitored, so on the label’s own logic it is not covered by that instruction. The only actionable item the labels generate is the one already built into statin care: a lipid panel.

What this leaves for a cash intake

Nothing above supports stopping a statin, and nothing supports adding a monitoring step that statin care does not already include. What it does support is narrower and easier to miss. A lipid panel drawn during the months of rapid weight loss will show triglycerides falling steeply and LDL cholesterol moving very little, and in a person with type 2 diabetes it may show LDL moving the wrong way. Read without that context, the same panel reads as a statin that has stopped working.

That reading belongs to whoever ordered the statin, which is the structural gap. A prescription written through a checkout does not produce a lipid panel, a comparison against the pre-treatment one, or a prescriber positioned to interpret the split — the pattern described in the telehealth article. The labeled instruction is addressed to a clinician holding both prescriptions, and the intake is where it stops being addressed to anyone.

One product distinction closes this out. Compounded semaglutide and tirzepatide are not FDA-approved, and the agency does not review them for safety, efficacy or quality before a pharmacy dispenses them; what that status does and does not mean is covered in a separate page. No compounded preparation appeared in any of the interaction studies above, and none arrives with the prescribing information that carries the oral-medication instruction at all. How claims here are established is set out in the methodology.

Frequently asked

Does a GLP-1 change how a statin is absorbed?
It changes how fast, and the direction is not consistent between products. Dulaglutide reduced atorvastatin exposure by 21%, subcutaneous semaglutide left it inside the prespecified 0.80 to 1.25 no-effect window, and oral semaglutide raised rosuvastatin exposure by 41% with a confidence interval of 1.24 to 1.60. All four published studies concluded no dose adjustment is required.
Why did a 41% exposure increase not produce a warning?
Because a statin dose is titrated against an LDL cholesterol target rather than a blood concentration, and it has a wide margin between an effective dose and a toxic one. A change of that size shows up as a lipid panel that has shifted, which is the measurement being taken anyway. That is why no statin appears in the drug interactions section of either weight-management label.
Does losing weight on a GLP-1 replace the need for a statin?
The label's own trial tables do not support that. At week 68, LDL cholesterol fell 3.8% relative to placebo in the obesity trial and 7.1% in the intensive-lifestyle trial, but rose 0.4% relative to placebo in the type 2 diabetes trial. Triglycerides fell between 13.9% and 17.0% across the same three trials, so the benefit lands almost entirely in a column a statin is not prescribed for.
Was the cardiovascular benefit shown in people who had stopped their statin?
No. In the 17,604-participant outcome trial, 90% of participants were taking lipid-lowering therapy at baseline, alongside 86% on a platelet aggregation inhibitor and 74% on an ACE inhibitor or angiotensin receptor blocker. The benefit was demonstrated as an addition to that background therapy, and stopping it was never tested.
If muscle aches start after adding a GLP-1, is the statin the cause?
A 2026 meta-analysis of 19 double-blind trials covering 123,940 participants found that only four of 66 further adverse outcomes listed in statin labels were significant after false-discovery control, and concluded that the blinded data do not support causal relationships for most listed effects. Fatigue and muscle symptoms during a dose escalation have a much better-documented source in the newly added drug.
What should actually be watched?
A lipid panel, which statin care already includes. The thing worth knowing before reading it is that the two columns move at very different speeds during rapid weight loss, so a panel showing triglycerides down sharply and LDL barely changed is the expected result rather than evidence that a statin has failed. Interpreting it requires whoever prescribed the statin to know the GLP-1 exists.

Sources

  1. [1] Novo Nordisk Pharmaceutical Industries, LP (2026). WEGOVY (semaglutide) injection and tablet — Drug Interactions 7.2, and Clinical Studies 14: Cardiometabolic Parameters DailyMed, U.S. National Library of Medicine. Source
  2. [2] Langeskov EK, Kristensen K (2022). Population pharmacokinetic of paracetamol and atorvastatin with co-administration of semaglutide. Pharmacol Res Perspect. PMID 35799471
  3. [3] Calvarysky B, Dotan I, Shepshelovich D, et al. (2024). Drug-Drug Interactions Between Glucagon-Like Peptide 1 Receptor Agonists and Oral Medications: A Systematic Review. Drug Saf. PMID 38273155
  4. [4] Hausner H, Derving Karsbøl J, Holst AG, et al. (2017). Effect of Semaglutide on the Pharmacokinetics of Metformin, Warfarin, Atorvastatin and Digoxin in Healthy Subjects. Clin Pharmacokinet. PMID 28349387
  5. [5] de la Peña A, Cui X, Geiser J, Loghin C (2017). No Dose Adjustment is Recommended for Digoxin, Warfarin, Atorvastatin or a Combination Oral Contraceptive When Coadministered with Dulaglutide. Clin Pharmacokinet. PMID 28357715
  6. [6] Jordy AB, Albayaty M, Breitschaft A, et al. (2021). Effect of Oral Semaglutide on the Pharmacokinetics of Levonorgestrel and Ethinylestradiol in Healthy Postmenopausal Women and Furosemide and Rosuvastatin in Healthy Subjects. Clin Pharmacokinet. PMID 33782832
  7. [7] Li F, Du C, Yu Q, et al. (2026). Effect of a Novel GLP-1 Analogue Ecnoglutide on the Pharmacokinetics of Rosuvastatin and Digoxin in Healthy Participants. Diabetes Obes Metab. PMID 42310888
  8. [8] Cholesterol Treatment Trialists' (CTT) Collaboration (2026). Assessment of adverse effects attributed to statin therapy in product labels: a meta-analysis of double-blind randomised controlled trials. Lancet. PMID 41655587
  9. [9] Eli Lilly and Company (2026). ZEPBOUND (tirzepatide) injection — Drug Interactions 7.2: Oral Medications DailyMed, U.S. National Library of Medicine. Source
  10. [10] Novo Nordisk Pharmaceutical Industries, LP (2026). OZEMPIC (semaglutide) injection — Clinical Pharmacology 12.3: Drug Interaction Studies DailyMed, U.S. National Library of Medicine. Source

Where to get it

Best GLP-1 injections

Every injectable seller we can verify, with the price each one publishes and an honest read of what the trials measured.

Compare providers →

More in Evidence